UTS site search

Professor CT Lin

Biography

Director, Computational Intelligence and Brain Computer Interface, FEIT, UTS

Dr. Chin-Teng Lin received the B.S. degree from National Chiao-Tung University (NCTU), Taiwan in 1986, and the Master and Ph.D. degree in electrical engineering from Purdue University, USA in 1989 and 1992, respectively. He is currently the Chair Professor of  Faculty of Engineering and Information Technology, University of Technology Sydney, Chair Professor of Electrical and Computer Engineering, NCTU, International Faculty of University of California at San-Diego (UCSD), and Honorary Professorship of University of Nottingham. Dr. Lin was elevated to be an IEEE Fellow for his contributions to biologically inspired information systems in 2005, and was elevated International Fuzzy Systems Association (IFSA) Fellow in 2012. He is elected as the Editor-in-chief of IEEE Transactions on Fuzzy Systems since 2011. He also served on the Board of Governors at IEEE Circuits and Systems (CAS) Society in 2005-2008, IEEE Systems, Man, Cybernetics (SMC) Society in 2003-2005, IEEE Computational Intelligence Society (CIS) in 2008-2010, and Chair of IEEE Taipei Section in 2009-2010. Dr. Lin is the Distinguished Lecturer of IEEE CAS Society from 2003 to 2005, and CIS Society from 2015-2017. He served as the Deputy Editor-in-Chief of IEEE Transactions on Circuits and Systems-II in 2006-2008. Dr. Lin was the Program Chair of IEEE International Conference on Systems, Man, and Cybernetics in 2005 and General Chair of 2011 IEEE International Conference on Fuzzy Systems. Dr. Lin is the coauthor of Neural Fuzzy Systems (Prentice-Hall), and the author of Neural Fuzzy Control Systems with Structure and Parameter Learning (World Scientific). He has published more than 200 journal papers (Total Citation: 20,155, H-index: 53, i10-index: 373) in the areas of neural networks, fuzzy systems, multimedia hardware/software, and cognitive neuro-engineering, including approximately 101 IEEE journal papers.

Professional

Fellow, Institute of Electrical and Electronic Engineers (IEEE) (2005)

Fellow, International Fuzzy Systems Association (IFSA) (2013)

Outstanding Achievement Award by Asia Pacific Neural Network Assembly (APNNA) (2013)

Outstanding Electrical and Computer Engineer (OECE) by School of Electrical and Computer Engineer of Purdue University (2011)

IEEE Distinguished Lecture Speaker (2003–2005, 2015–2017)

Editor-in-Chief, IEEE Transactions on Fuzzy Systems (2011–2016)

Founding Editor-in-Chief, Journal of Neuroscience and Neuroengineering, American Scientific Publishers (2012–)

Deputy Editor-in-Chief, IEEE Trans. on Circuits and Systems II (2006–07)

Chairman, IEEE Taipei Section (2009-2010) (received the IEEE Member and Geographic Activities Board Outstanding Large Section Award)

President, Asia Pacific Neural Network Assembly (APNNA) (2004-2005)

General Chair, 2011 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2011), Taipei, Taiwan, 2011.

Program co-chairs, 2015 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2015), Istanbul, Turkey, 2015.

Panel Sessions chair, 2016 The IEEE World Congress on Computational Intelligence (IEEE WCCI 2016), Vancouver, Canada, 2016.

Panel Sessions Chair, 2014 IEEE World Congress on Computational Intelligence (WCCI 2014), Beijing, China, 2014.

Invited Session Chair, 2012 IEEE World Congress on Computational Intelligence (IEEE-WCCI 2012), Brisbane, Australia, 2012.

Merit NSC Research Fellow Award, National Science Council (NSC), Taiwan, 2009.

Outstanding Research Award, National Science Council (NSC), Taiwan, 1996 – Present.

The 38th Ten Outstanding Rising Stars in Taiwan granted by International Junior Chamber – Taiwan Chapter, 2000.

Taiwan Outstanding Information-Technology Expert Award, 2002.

Outstanding Electrical Engineering Professor Award, the Chinese Institute of Electrical Engineering (CIEE), 1997.

Outstanding Engineering Professor Award, the Chinese Institute of Engineering (CIE), 2000.

Image of CT Lin
Distinguished Professor, School of Software
PhD
 
Phone
+61 2 9514 1687

Research Interests

My current research interests include the following domains:
Computational intelligence, fuzzy neural networks (FNN), cognitive neuro-engineering, brain-computer interface, multimedia information processing, machine learning, robotics, and intelligent sensing and control.


Important Academic Achievements:
Publication: (Total Citation: ~19,166, H-index: 53, i10-index: 332)
Journal papers: over 200 (including 101 IEEE Transactions/magazine papers)
Conference papers: over 250
Patents: 124
I devoted myself to the research of Computational Intelligence (CI) since her infant stage of early 1990’s. I took the lead in developing the models of fuzzy neural networks (FNN), which have now become one of the most prevailing research areas in CI. Since 1996, the Intelligent Information Technology (iIT) has been my major research interest.  In 2003, I founded the Brain Research Center (BRC) at the National Chao-Tung University (NCTU) and served as its director to this day. By applying the CI and iIT to the research of cognition neuroscience, I have focused mainly on the area of Translational Neuroscience since 2003, trying to bring basic findings in neuroscience into daily life applications. To achieve this goal, I have focused on 2 major problems: Natural Cognition and Brain-Computer Interface (BCI). Natural cognition studies the brain and its behavior at work. One major research problem is the study of brain dynamics of alertness, drowsiness, motional sickness, distraction, orientation, etc. during driving. For real-world applications of BCI, we have designed wearable and wireless devices for measuring brain waves (EEG signals) with dry sensors and developed algorithms for brain waves analysis. The Brain Research Center is now a platform for conducting various interdisciplinary research activities in cognition neuroscience and engineering. 

Can supervise: Yes
Machine Learning, Logic Design, Fuzzy Systems, Neural Networks, Computational Intelligence, Cognitive Neuroscience

Chapters

Lin, C.T., Chen, C.H. & Lin, C.J. 2009, 'Nonlinear system control using functional-link-based neuro-fuzzy networks' in Recent Advances in Intelligent Control Systems, pp. 249-275.
View/Download from: Publisher's site
This study presents a functional-link-based neuro-fuzzy network (FLNFN) structure for nonlinear system control. The proposed FLNFN model uses a functional link neural network (FLNN) to the consequent part of the fuzzy rules. This study uses orthogonal polynomials and linearly independent functions in a functional expansion of the FLNN. Thus, the consequent part of the proposed FLNFN model is a nonlinear combination of input variables. An online learning algorithm, which consists of structure learning and parameter learning, is also presented. The structure learning depends on the entropy measure to determine the number of fuzzy rules. The parameter learning, based on the gradient descent method, can adjust the shape of the membership function and the corresponding weights of the FLNN. Finally, the FLNFN model is applied in various simulations. Results of this study demonstrate the effectiveness of the proposed FLNFN model. © 2009 Springer London.

Conferences

Singh, A., Wang, Y.-.K., Chiu, C.-.Y., Yu, Y.-.H., Nascimben, M., King, J.-.T., Chuang, C.-.H., Chen, S.-.A., Ko, L.-.W., Pal, N.R. & others 2016, 'Attention in Complex Environment of Brain Computer Interface', Proceedings of the 6th International Brain-Computer Interface Meeting, organized by the BCI Society, Verlag der TU Graz, Graz University of Technology, sponsored by g.tec medical engineering GmbH, pp. 165-165.
Nascimben, M., Yu, Y.-.H., Lin, C.-.T., King, J.-.T., Singh, A.K. & Chuang, C.-.H. 2016, 'Effect of a cognitive involving videogame on MI task', Proceedings of the 6th International Brain-Computer Interface Meeting, organized by the BCI Society, Verlag der TU Graz, Graz University of Technology, sponsored by g.tec medical engineering GmbH, pp. 175-175.
Liu, Y.T., Wu, S.L., Chou, K.P., Lin, Y.Y., Lu, J., Zhang, G., Lin, W.C. & Lin, C.T. 2016, 'Driving fatigue prediction with pre-event electroencephalography (EEG) via a recurrent fuzzy neural network', 2016 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2016, pp. 2488-2494.
View/Download from: Publisher's site
© 2016 IEEE.We propose an electroencephalography (EEG) prediction system based on a recurrent fuzzy neural network (RFNN) architecture to assess drivers' fatigue degrees during a virtual-reality (VR) dynamic driving environment. Prediction of fatigue degrees is a crucial and arduous biomedical issue for driving safety, which has attracted growing attention of the research community in the recent past. Meanwhile, combined with the benefits of measuring EEG signals facilitates, many EEG-based brain-computer interfaces (BCIs) have been developed for use in real-Time mental assessment. In the literature, EEG signals are severely blended with stochastic noise; therefore, the performance of BCIs is constrained by low resolution in recognition tasks. For this rationale, independent component analysis (ICA) is usually used to find a source mapping from original data that has been blended with unrelated artificial noise. However, the mechanism of ICA cannot be used in real-Time BCI design. To overcome this bottleneck, the proposed system in this paper utilizes a recurrent self-evolving fuzzy neural work (RSEFNN) to increase memory capability for adaptive noise cancellation when assessing drivers' mental states during a car driving task. The experimental results without the use of ICA procedure indicate that the proposed RSEFNN model remains superior performance compared with the state-of-Thearts models.
Liu, Y.T., Wu, S.L., Chou, K.P., Lin, Y.Y., Lu, J., Zhang, G., Chuang, C.H., Lin, W.C. & Lin, C.T. 2016, 'A motor imagery based brain-computer interface system via swarm-optimized fuzzy integral and its application', 2016 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2016, pp. 2495-2500.
View/Download from: Publisher's site
© 2016 IEEE.A brain-computer interface (BCI) system provides a convenient means of communication between the human brain and a computer, which is applied not only to healthy people but also for people that suffer from motor neuron diseases (MNDs). Motor imagery (MI) is one well-known basis for designing Electroencephalography (EEG)-based real-life BCI systems. However, EEG signals are often contaminated with severe noise and various uncertainties, imprecise and incomplete information streams. Therefore, this study proposes spectrum ensemble based on swam-optimized fuzzy integral for integrating decisions from sub-band classifiers that are established by a sub-band common spatial pattern (SBCSP) method. Firstly, the SBCSP effectively extracts features from EEG signals, and thereby the multiple linear discriminant analysis (MLDA) is employed during a MI classification task. Subsequently, particle swarm optimization (PSO) is used to regulate the subject-specific parameters for assigning optimal confidence levels for classifiers used in the fuzzy integral during the fuzzy fusion stage of the proposed system. Moreover, BCI systems usually tend to have complex architectures, be bulky in size, and require time-consuming processing. To overcome this drawback, a wireless and wearable EEG measurement system is investigated in this study. Finally, in our experimental result, the proposed system is found to produce significant improvement in terms of the receiver operating characteristic (ROC) curve. Furthermore, we demonstrate that a robotic arm can be reliably controlled using the proposed BCI system. This paper presents novel insights regarding the possibility of using the proposed MI-based BCI system in real-life applications.
Liu, A., Zhang, G., Lu, J., Lu, N. & Lin, C.T. 2016, 'An online competence-based concept drift detection algorithm', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 416-428.
View/Download from: Publisher's site
© Springer International Publishing AG 2016.The ability to adapt to new learning environments is a vital feature of contemporary case-based reasoning system. It is imperative that decision makers know when and how to discard outdated cases and apply new cases to perform smart maintenance operations. Competencebased empirical distance has been recently proposed as a measurement that can estimate the difference between case sample sets without knowing the actual case distributions. It is reportedly one of the most accurate drift detection algorithms in both synthetic and real-world data sets. However, as the construction of competence models have to retain every case in memory, it is not suitable for online drift detection. In addition, the high computational complexity O(n2) also limits its practical application, especially when dealing with large scale data sets with time constrains. In this paper, therefore, we propose a space-based online case grouping strategy, and a new case group enhanced competence distance (CGCD), to address these issues. The experiment results show that the proposed strategy and related algorithms significantly improve the efficiency of the current leading competence-based drift detection algorithm.
Gu, F., Zhang, G., Lu, J. & Lin, C.T. 2016, 'Concept drift detection based on equal density estimation', Proceedings of the International Joint Conference on Neural Networks, pp. 24-30.
View/Download from: Publisher's site
© 2016 IEEE.An important problem that remains in online data mining systems is how to accurately and efficiently detect changes in the underlying distribution of large data streams. The challenge for change detection methods is to maximise the accumulative effect of changing regions with unknown distribution, while at the same time providing sufficient information to describe the nature of the changes. In this paper, we propose a novel change detection method based on the estimation of equal density regions, with the aim of overcoming the issues of instability and inefficiency that underlie methods of predefined space partitioning schemes. Our method is general, nonparametric and requires no prior knowledge of the data distribution. A series of experiments demonstrate that our method effectively detects concept drift in single dimension as well as high dimension data, and is also able to explain the change by locating the data points that contribute most to the change. The detection result is guaranteed by statistical tests.
Lin, C., Mao, Z.J., Jung, T.P. & Huang, Y.F. 2016, 'Predicting EEG Sample Size Required for Classification Calibration'.
Ko, L.W., Yang, B.J., Singanamalla, S.K.R., Lin, C., King, J.T. & Jung, T.P. 2016, 'A practical neurogaming design based on SSVEP brain computer interface'.
Ko, L.W., Komarov, S.H., Liu, S.H., Hsu, W.C., König, P., Goeke, P., David Hairston, W., Lin, C. & Jung, T.P. 2016, 'Investigation of brain activity patterns related to the effect of classroom fatigue'.
Davis, J.J., Kozma, R., Lin, C. & Freeman, W.J. 2016, 'Spatio-temporal EEG pattern extraction using high-density scalp arrays'.
Chang, C.L., Huang, C.S., Lu, S.W. & Lin, C. 2016, 'Real-Time Unsupervised Artifact Removal Algorithm Using Wearable Dry EEG System'.
Chang, C.L., Huang, C.S., Lu, S.W. & Lin, C. 2016, 'Apply Artifact Rejection on Multi-Channel Dry EEG System under Motion'.
Wu, D., Lawhern, V., Gordon, S., Lance, B. & Lin, C. 2016, 'Spectral Meta-Learner for Regression (SMLR) Model Aggregation: Towards Calibrationless Brain-Computer Interface'.
Wu, D., Lawhern, V., Gordon, S., Lance, B. & Lin, C. 2016, 'Agreement Rate Initialized Maximum Likelihood Estimator (ARIMLE) for Ensemble Classifier Aggregation and Its Application in Brain-Computer Interface'.
Nascimben, M., King, J.T. & Lin, C.T. 2016, 'Resting Upper Alpha Can Predict Motor Imagery Performance?'.
Wang, F., He, Y.B., Qu, J., Xie, Q.Y., Lin, Q., Ni, X.X., Chen, Y., Yu, R.H. & Lin, C.T. 2016, 'An Audiovisual BCI System for Assisting Clinical Communication Assessment in Patients with Disorders of Consciousness: A Case Study'.
Wei, C.S., Lin, Y.P., Wang, Y.T., Lin, C. & Jung, T.P. 2016, 'Transfer Learning with Large-Scale Data in Brain-Computer Interfaces'.
Wu, D., Lawhern, V., Gordon, S., Lance, B. & Lin, C. 2016, 'Offline EEG-based driver drowsiness estimation using enhanced batch-mode active learning (EBMAL) for regression'.
Liu, Y.T., Pal, N.R., Wu, S.L., Hsieh, T.Y. & Lin, C. 2016, 'Adaptive Subspace Sampling for Class Imbalance Processing'.
Chen, H., Zhang, G., Lu, J. & Zhu, D. 2015, 'A Fuzzy Approach for Measuring Development of Topics in Patents Using Latent Dirichlet Allocation', 2015 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2015), The 2015 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2015), IEEE, Istanbul, Turkey, pp. 1-7.
View/Download from: UTS OPUS or Publisher's site
Technology progress brings the very rapid growth of patent publications, which increases the difficulty of domain experts to measure the development of various topics, handle linguistic terms used in evaluation and understand massive technological content. To overcome the limitations of keyword-ranking type of text mining result in existing research, and at the same time deal with the vagueness of linguistic terms to assist thematic evaluation, this research proposes a fuzzy set-based topic development measurement (FTDM) approach to estimate and evaluate the topics hidden in a large volume of patent claims using Latent Dirichlet Allocation. In this study, latent semantic topics are first discovered from patent corpus and measured by a temporal-weight matrix to reveal the importance of all topics in different years. For each topic, we then calculate a temporal-weight coefficient based on the matrix, which is associated with a set of linguistic terms to describe its development state over time. After choosing a suitable linguistic term set, fuzzy membership functions are created for each term. The temporal-weight coefficients are then transformed to membership vectors related to the linguistic terms, which can be used to measure the development states of all topics directly and effectively. A case study using solar cell related patents is given to show the effectiveness of the proposed FTDM approach and its applicability for estimating hidden topics and measuring their corresponding development states efficiently.
Lin, C.T., Wang, Y.K., Fang, C.N., Yu, Y.H. & King, J.T. 2015, 'Extracting patterns of single-trial EEG using an adaptive learning algorithm', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 6642-6645.
View/Download from: Publisher's site
© 2015 IEEE. The improvement of brain imaging technique brings about an opportunity for developing and investigating brain-computer interface (BCI) which is a way to interact with computer and environment. The measured brain activities usually constitute the signals of interest and noises. Applying the portable device and removing noise are the benefits to real-world BCI. In this study, one portable electroencephalogram (EEG) system non-invasively acquired brain dynamics through wireless transmission while six subjects participated in the rapid serial visual presentation (RSVP) paradigm. The event-related potential (ERP) was traditionally estimated by ensemble averaging (EA) to increase the signal-to-noise ratio. One adaptive filter of data-reusing radial basis function network (DR-RBFN) was also utilized as the estimator. The results showed that this portable EEG system stably acquired brain activities. Furthermore, the task-related potentials could be clearly explored from the limited samples of EEG data through DR-RBFN. According to the artifact-free data from the portable device, this study demonstrated the potential to move the BCI from laboratory research to real-life application in the near future.
Cao, Z.H., Ko, L.W., Lai, K.L., Huang, S.B., Wang, S.J. & Lin, C.T. 2015, 'Classification of migraine stages based on resting-state EEG power', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: UTS OPUS or Publisher's site
© 2015 IEEE. Migraine is a chronic neurological disease characterized by recurrent moderate to severe headaches during a period like one month often in association with symptoms in human brain and autonomic nervous system. Normally, migraine symptoms can be categorized into four different stages: inter-ictal, pre-ictal, ictal, and post-ictal stages. Since migraine patients are difficulty knowing when they will suffer migraine attacks, therefore, early detection becomes an important issue, especially for low-frequency migraine patients who have less than 5 times attacks per month. The main goal of this study is to develop a migraine-stage classification system based on migraineurs' resting-state EEG power. We collect migraineurs' O1 and O2 EEG activities during closing eyes from occipital lobe to identify pre-ictal and non-pre-ictal stages. Self-Constructing Neural Fuzzy Inference Network (SONFIN) is adopted as the classifier in the migraine stages classification which can reach the better classification accuracy (66%) in comparison with other classifiers. The proposed system is helpful for migraineurs to obtain better treatment at the right time.
Ko, L.W., Lai, W.K., Liang, W.G., Chuang, C.H., Lu, S.W., Lu, Y.C., Hsiung, T.Y., Wu, H.H. & Lin, C.T. 2015, 'Single channel wireless EEG device for real-time fatigue level detection', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: UTS OPUS or Publisher's site
© 2015 IEEE. Driver fatigue problem is one of the important factors of traffic accidents. Recent years, many research had investigated that using EEG signals can effectively detect driver's drowsiness level. However, real-time monitoring system is required to apply these fatigue level detection techniques in the practical application, especially in the real-road driving. Therefore, it required less channels, portable and wireless, real-time monitoring and processing techniques for developing the real-time monitoring system. In this study, we develop a single channel wireless EEG device which can real-time detect driver's fatigue level on the mobile device such as smart phone or tablet. The developed device is investigated to obtain a better and precise understanding of brain activities of mental fatigue under driving, which is of great benefit for devolvement of detection of driving fatigue system. This system consists of a Bluetooth-enabled one channel EEG, a regression model, and smartphone, which was a platform recording and transforming the raw EEG data to useful driving status. In the experiment, this was a sustained-attention driving task to implement in a virtual-reality (VR) driving simulator. To training model and develop the system, we were performed for 15 subjects to study Electroencephalography (EEG) brain dynamics by using a mobile and wireless EEG device. Based on the outstanding training results, the leave-one-subject-out cross validation test obtained 90% fatigue detection accuracy. These results indicate that the combination of a smartphone and wireless EEG device constitutes an effective and easy wearable solution for detecting and preventing driver fatigue in real driving environments.
Liu, Y.T., Lin, Y.Y., Hsieh, T.Y., Wu, S.L. & Lin, C.T. 2015, 'A global optimized neuro-fuzzy system using artificial bee colony evolutionary algorithm', Frontiers in Artificial Intelligence and Applications, pp. 140-149.
View/Download from: Publisher's site
© 2015 The authors and IOS Press. All rights reserved. This paper proposes a novel Takagi-Sugeno-Kang-type (TSK-type) neuro-fuzzy system (NFS), which utilizes the artificial bee colony (ABC) evolutionary algorithm for parameter optimization. The ABC evolutionary algorithm was developed based on imitating foraging behavior of natural bees for numerical optimization problems, and it has been proved to outperform other metaheuristic approaches on different constrain optimization problems in previous studies. The proposed NFS in this paper adopts an adaptive clustering method to generate fuzzy rules for determining the system architecture, and the TSK-type reasoning is employed for the consequent part of each rule. Subsequently, all free parameters in the NFS designed, including the premise and the consequent parameters, will be optimized by ABC algorithm. This study compares the performance of ABC algorithm with that of Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Differential Evolution (DE). The simulation results show that the performance of ABC algorithm is superior to that of the mentioned algorithms for solving dynamic system problems.
Chiu, C.Y., Chen, C.Y., Lin, Y.Y., Chen, S.A. & Lin, C.T. 2014, 'Using a novel LDA-ensemble framework to classification of motor imagery tasks for brain-computer interface applications', Frontiers in Artificial Intelligence and Applications, International Computer Symposium (ICS 2014), IOS Press, Tunghai University, Taichung, Taiwan, pp. 150-156.
View/Download from: Publisher's site
© 2015 The authors and IOS Press. All rights reserved. In this paper, we introduce a novel linear discriminate analysis (LDA) ensemble classifier utilizing the Mindo as a brain-computer interface (BCI) device to deal with the problem of motor imagery classification. With regard to the composition of the proposed system, we combine filter bank, sub-band common spatial pattern (SBCSP), LDA together for extracting features of EEG data and classifying the motor imagery with left or right states. In addition, we also employ a gradient descent (GD) algorithm to find the best weight associated with probability fusion function. This novel architecture not only boosts the accuracy of classification but maintains the computational efficiency of the system. Therefore, the proposed LDA-ensemble framework is able to be satisfied with each subject as demonstrated in Section III.
Chou, K.P., Prasad, M., Lin, Y.Y., Joshi, S., Lin, C.T. & Chang, J.Y. 2015, 'Takagi-Sugeno-Kang type collaborative fuzzy rule based system', IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CIDM 2014: 2014 IEEE Symposium on Computational Intelligence and Data Mining, Proceedings, pp. 315-320.
View/Download from: Publisher's site
© 2014 IEEE. In this paper, a Takagi-Sugeno-Kang (TSK) type collaborative fuzzy rule based system is proposed with the help of knowledge learning ability of collaborative fuzzy clustering (CFC). The proposed method split a huge dataset into several small datasets and applying collaborative mechanism to interact each other and this process could be helpful to solve the big data issue. The proposed method applies the collective knowledge of CFC as input variables and the consequent part is a linear combination of the input variables. Through the intensive experimental tests on prediction problem, the performance of the proposed method is as higher as other methods. The proposed method only uses one half information of given dataset for training process and provide an accurate modeling platform while other methods use whole information of given dataset for training.
Prasad, M., Chou, K.P., Saxena, A., Kawrtiya, O.P., Li, D.L. & Lin, C.T. 2015, 'Collaborative fuzzy rule learning for Mamdani type fuzzy inference system with mapping of cluster centers', IEEE SSCI 2014 - 2014 IEEE Symposium Series on Computational Intelligence - CICA 2014: 2014 IEEE Symposium on Computational Intelligence in Control and Automation, Proceedings.
View/Download from: Publisher's site
© 2014 IEEE. This paper demonstrates a novel model for Mamdani type fuzzy inference system by using the knowledge learning ability of collaborative fuzzy clustering and rule learning capability of FCM. The collaboration process finds consistency between different datasets, these datasets can be generated at various places or same place with diverse environment containing common features space and bring together to find common features within them. For any kind of collaboration or integration of datasets, there is a need of keeping privacy and security at some level. By using collaboration process, it helps fuzzy inference system to define the accurate numbers of rules for structure learning and keeps the performance of system at satisfactory level while preserving the privacy and security of given datasets.
Wu, D., Chuang, C.H. & Lin, C.T. 2015, 'Online driver's drowsiness estimation using domain adaptation with model fusion', 2015 International Conference on Affective Computing and Intelligent Interaction, ACII 2015, 2015 International Conference on Affective Computing and Intelligent Interaction (ACII), IEEE, Piscataway, USA, pp. 904-910.
View/Download from: Publisher's site
© 2015 IEEE. Drowsy driving is a pervasive problem among drivers, and is also an important contributor to motor vehicle accidents. It is very important to be able to estimate a driver's drowsiness level online so that preventative actions could be taken to avoid accidents. However, because of large individual differences, it is very challenging to design an estimation algorithm whose parameters fit all subjects. Some subject-specific calibration data must be used to tailor the algorithm for each new subject. This paper proposes a domain adaptation with model fusion (DAMF) online drowsiness estimation approach using EEG signals. By making use of EEG data from other subjects in a transfer learning framework, DAMF requires very little subject-specific calibration data, which significantly increases its utility in practice. We demonstrate using a simulated driving experiment and 15 subjects that DAMF can achieve much better performance than several other approaches.
Singh, A.K., Wang, Y.-.K., King, J.-.T., Lin, C.-.T. & Ko, L.-.W. 2015, 'A simple communication system based on Brain Computer Interface', Technologies and Applications of Artificial Intelligence (TAAI).
Chen, S.A., Chen, C.H., Lin, J.W., Ko, L.W. & Lin, C.T. 2014, 'Gaming controlling via brain-computer interface using multiple physiological signals', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 3156-3159.
View/Download from: Publisher's site
© 2014 IEEE. using physiological signals to control braincomputer interface (BCI) becomes more popular. Among many kinds of physiological signals, Electrooculography (EOG) signal is more stable which can be used to control BCI systems based on eye movement detection and signal processing methods. Also, the use of electroencephalographic (EEG) signals has become the most common approach for a BCI because of their usability and strong reliability. In this paper, we described a signal processing method, which uses a wireless EEG-based BCI system designed to be worn near forehead that can detect both EEG and EOG signals, for detecting eye movements to have 9 direction controls (via EOG) and one action of execution (via EEG). This system included a wireless EEG signal acquisition device, a mechanism that can be worn stably, and an application program (APP) with signal processing algorithms. This algorithm and its classification procedure provided an effective method for identifying eye movements and attention. Finally, we designed a baseball game to test the BCI system. The results demonstrated that player can control the game well with high accuracy.
Zao, J.K., Lin, C.T., Ko, L.W., She, H.C., Dung, L.R. & Chen, B.Y. 2014, 'Natural user interfaces: Cyber-physical challenges and pervasive applications - A panel discussion', Proceedings - 2014 IEEE International Conference on Internet of Things, iThings 2014, 2014 IEEE International Conference on Green Computing and Communications, GreenCom 2014 and 2014 IEEE International Conference on Cyber-Physical-Social Computing, CPS 2014, pp. 467-469.
View/Download from: Publisher's site
© 2014 IEEE. This panel discussion aims at exploring the potential applications of emerging natural user interface (NUI) technologies and the challenges they pose to the design and deployment of cyber-physical systems. Based on their research work, six panelists will take turns to present the outlook, the cyber-physical requirements and the promising applications of implicit NUI. We hope these short presentations will lead to thought-provoking discussions and inspire further innovation.
Prasad, M., Siana, L., Li, D.L., Lin, C.T., Liu, Y.T. & Saxena, A. 2014, 'A preprocessed induced partition matrix based collaborative fuzzy clustering for data analysis', IEEE International Conference on Fuzzy Systems, pp. 1553-1558.
View/Download from: Publisher's site
© 2014 IEEE. Preprocessing is generally used for data analysis in the real world datasets that are noisy, incomplete and inconsistent. In this paper, preprocessing is used to refine the inconsistency of the prototype and partition matrices before getting involved in the collaboration process. To date, almost all organizations are trying to establish some collaboration with others in order to enhance the performance of their services. Due to privacy and security issues they cannot share their information and data with each other. Collaborative clustering helps this kind of collaborative process while maintaining the privacy and security of data and can still yield a satisfactory result. Preprocessing helps the collaborative process by using an induced partition matrix generated based on cluster prototypes. The induced partition matrix is calculated from local data by using the cluster prototypes obtained from other data sites. Each member of the collaborating team collects the data and generates information locally by using the fuzzy c-means (FCM) and shares the cluster prototypes to other members. The other members preprocess the centroids before collaboration and use this information to share globally through collaborative fuzzy clustering (CFC) with other data. This process helps system to learn and gather information from other data sets. It is found that preprocessing helps system to provide reliable and satisfactory result, which can be easily visualized through our simulation results in this paper.
Wu, S.L., Lin, Y.Y., Liu, Y.T., Chen, C.Y. & Lin, C.T. 2014, 'A learning scheme to Fuzzy C-Means based on a compromise in updating membership degrees', IEEE International Conference on Fuzzy Systems, pp. 1534-1537.
View/Download from: Publisher's site
© 2014 IEEE. Fuzzy C-Means (FCM) clustering is the most well-known clustering method according to fuzzy partition for pattern classification. However, there are some disadvantages of using that clustering method, such as computational complexity and execution time. Therefore, to solve these drawbacks of FCM, the two-phase FCM procedure has been proposed in this study. Compared with the conventional FCM, the usage of a compromised learning scheme makes more adaptive and effective. By performing the proposed approach, the unknown data could be rapidly clustered according to the previous information. A synthetic data set with two dimensional variables is generated to estimate the performance of the proposed method, and to further demonstrate that our method not only reduces computational complexity but economizes execution time compared with the conventional FCM in each example.
Liu, Y.T., Lin, Y.Y., Wu, S.L., Chuang, C.H., Prasad, M. & Lin, C.T. 2014, 'EEG-based driving fatigue prediction system using functional-link-based fuzzy neural network', Proceedings of the International Joint Conference on Neural Networks, pp. 4109-4113.
View/Download from: Publisher's site
© 2014 IEEE. This study presents a fuzzy prediction system for the forecasting and estimation of driving fatigue, which utilizes a functional-link-based fuzzy neural network (FLFNN) to predict the drowsiness (DS) level in car driving task. The cognitive state in car driving task is one of key issue in cognitive neuroscience because fatigue driving usually causes enormous losses nowadays. The damage can be extremely decreased by the assistant of various artificial systems. Many Electroencephalography (EEG)-based interfaces have been widely developed recently due to its convenient measurement and real-time response. However, the improvement of recognition accuracy is still confined to some specific problems (e.g., individual difference). In order to solve this issue, the proposed methodology in this paper utilizes a nonlinear fuzzy neural network structure to increase the adaptability in the real-world environment. Therefore, this study is further to analysis the brain activities in car driving, which is constructed in a simulated three-dimensional virtual-reality (VR) environment. Finally, through the development of brain cognitive model in car driving task, this system can predict the cognitive state effectively before drivers' action and then provide correct feedback to users. This study also compared the result with the-state-of-art systems, including Linear Regression (LR), Multi-Layer Perceptron Neural Network (MLPNN) and Support Vector Regression (SVR). Results of this study demonstrate the effectiveness of the proposed FLFNN model.
Chen, C.Y., Wu, C.W., Lin, C.T. & Chen, S.A. 2014, 'A novel classification method for motor imagery based on Brain-Computer Interface', Proceedings of the International Joint Conference on Neural Networks, pp. 4099-4102.
View/Download from: Publisher's site
© 2014 IEEE. Brain computer interface (BCI) is known as a good way to communicate between brain and computer or other device. There are many kinds of physiological signal can operate BCI systems. Motor imagery (MI) has been demonstrated to be a good way to operate a BCI system. In some recent studies about MI based BCI systems, low accuracy rate and time consuming are common problems. In this thesis, a novel motor imagery algorithm is proposed to improve the accuracy rate and computational efficiency at the same time. The architecture of many BCI system is quite complex and they involve time consuming processing. The electroencephalography (EEG) signal is the most commonly used inputs for BCI applications but EEG is often contaminated with noise. To overcome such drawbacks, in this paper we use the common spatial pattern (CSP) for feature extraction from EEG and the linear discriminant analysis (LDA) for motor imagery classification. In this study, CSP and LDA have been used to reduce the artifact and classify Mi-based EEG signal. We have used two-level cross validation scheme to determine the subject specific best time window and number of CSP features. We have compared the performance of our system with BCI competition results. This novel algorithm with high accuracy rate and efficiency can be applied to real time BCI system in real-life applications.
Lin, C.T., Prasad, M. & Chang, J.Y. 2013, 'Designing Mamdani type fuzzy rule using a collaborative FCM scheme', iFUZZY 2013 - 2013 International Conference on Fuzzy Theory and Its Applications, pp. 279-282.
View/Download from: Publisher's site
This paper presents a new approach for generating fuzzy rules for fuzzy inference system by using collaborative fuzzy c-mean (CFCM). In order to do any mode of integration between datasets, there is a need to define the common feature between datasets by using some kind of collaborative process and also need to preserve the privacy and security at higher levels. This collaboration process gives a common structure between datasets which helps to define an appropriate number of rules for structural learning and also improve the accuracy of the system modeling. This all consideration bring the concept of collaborative fuzzy rule generation process with a quality measuring. © 2013 IEEE.
Liao, S.H., Chang, J.Y. & Lin, C.T. 2013, 'Study on least trimmed absolute deviations artificial neural network', iFUZZY 2013 - 2013 International Conference on Fuzzy Theory and Its Applications, pp. 156-160.
View/Download from: Publisher's site
In this paper, the least trimmed sum of absolute deviations (LTA) estimator, frequently used in robust linear parametric regression problems, will be generalized to nonparametric least trimmed sum of absolute deviations-artificial neural network (LTA-ANN) for nonlinear regression problems. In linear parametric regression problems, the LTA estimator usually have good robustness against outliers and can theoretically tolerate up to 50% of outlying data. Moreover, a nonderivative hybrid method mixing the simplex method of Nelder and Mead (NM) and particle swarm optimization algorithm (PSO), abbreviated as SNM-PSO, will be provided in this study for the training of the parameters of LTA-ANN. Some numerical examples will be provided to compare the robustness against outliers for usual artificial neural network (ANN) and the proposed LTA-ANN. Simulation results show that the LTA-ANN proposed in this paper have good robustness against outliers. © 2013 IEEE.
Lin, C.T., Wu, S.L., Jiang, W.L., Liang, J.W. & Chen, S.A. 2013, 'A wireless Electrooculography-based human-computer interface for baseball game', ICICS 2013 - Conference Guide of the 9th International Conference on Information, Communications and Signal Processing.
View/Download from: Publisher's site
Gaming control becomes popular based on the development of human-computer interface (HCI). Among many kinds of physiological signals, Electrooculography (EOG) signal is more stable which can be used to control HCI systems based on eye movement detection and signal processing methods. However, there are currently no effective multi-directional classification methods for monitoring eye movements. In addition, many EOG-based HCI systems have been developed with traditional wet electrodes. Those traditional electrodes require conductive gel and skin preparation on some users. Here, we describe a signal processing method used in a wireless EOG-based HCI system with dry electrodes for detecting eye movements to have 9 options. This system includes wireless EOG signal acquisition device, dry electrodes and an EOG signal processing algorithm. The EOG signal processing algorithm is based on 9 options of eye movement and blink signals. The results demonstrated an application of baseball game control using the proposed wireless HCI system. This system provides an effective and convenient method for eye movement detection. © 2013 IEEE.
Ko, L.W., Lai, K.L., Huang, P.H., Lin, C.T. & Wang, S.J. 2013, 'Steady-state visual evoked potential based classification system for detecting migraine seizures', International IEEE/EMBS Conference on Neural Engineering, NER, pp. 1299-1302.
View/Download from: Publisher's site
The recurrent migraine attacks between interictal phenomenon is triggered by the migraineurs' brain lacking for habituation, due to the stimulations from the outside world that increase the excitability of brain activity, which have been considered as the possible reasons for migraine seizure. The variation of habituation level within the migraine cycle is proposed to be a critical symptom to describe the physiological states of migraine headache. This study proposed Steady-State Visual Evoked Potentials (SSVEP) examination to utilize habituation for classifying the different physiologic states of migraine cycle, and implement a classification system to determine different migraine states. The developed system may be extended to detect migraine seizure, and provide an opportunity to a clinically individual-based headache monitoring program, aiming for early migraine detection. © 2013 IEEE.
Khoo, I.H., Reddy, H.C., Van, L.D. & Lin, C.T. 2013, 'Design of 2-D digital filters with almost quadrantal symmetric magnitude response without 1-D separable denominator factor constraint', Midwest Symposium on Circuits and Systems, pp. 999-1002.
View/Download from: Publisher's site
A design approach is presented for 2-D digital filters possessing approximate quadrantal magnitude symmetry without the constraint of the denominator having only 1-D separable factors. To ensure the BIBO stability of the filter, the planar least square inverse stabilization approach is employed. It is illustrated through design examples that the proposed approach results in filters with sharper transition band and lower error relative to the given filter specifications. Also, for certain cases, it is shown that a lower order non-separable denominator design can achieve the same result as a higher order separable denominator design, thus providing savings in the number of multipliers. Finally, 2-D VLSI realizations without global broadcast are presented for the optimized transfer function with non-separable denominator factors and approximate quadrantal symmetry. © 2013 IEEE.
Huang, C.S., Lin, C.L., Ko, L.W., Wang, Y.K., Liang, J.W. & Lin, C.T. 2013, 'Automatic sleep stage classification GUI with a portable EEG device', Communications in Computer and Information Science, pp. 613-617.
View/Download from: Publisher's site
In this study, a developed automatic sleep stage classification system with a portable EEG recording device, (Mindo-4s) is implemented by JAVA-based sleep graphical user interface (GUI) in android platform. First, the parameters of the developed sleep stage classification system, including extracting effective sleep features and a hierarchical classification structure consisting of preliminary wake detection rule, adaptive adjustment scheme, and support vector machine, were trained by our existing sleep database, which collected using polysomnogram (PSG), in MATLAB program. Finally, this classification system would be reedited by JAVA language, and the corresponding JAVA-based sleep GUI software was working in android platform and Mindo-4s. The connection between JAVA-based sleep GUI software and the portable Mindo-4s was through Bluetooth communication. The performance of this JAVA-based sleep GUI can reach 72.43% average accuracy comparing to the result from manual scoring. This JAVA-based sleep GUI can on-line display, record and analyze the forehead EEG signals simultaneously. After sleep, the user can received a complete sleep report, including sleep efficiency, sleep stage distribution, from JAVA-based sleep GUI. Thus, this system can provide a preliminary result in sleep quality estimation, and help the sleep doctor to decide someone needs to have a complete PSG testing in hospital. Using this system is more convenient for long-term and home-based daily caring than traditional PSG measurement. © Springer-Verlag Berlin Heidelberg 2013.
Wang, Y.K., Jung, T.P., Chen, S.A., Huang, C.S. & Lin, C.T. 2013, 'Tracking attention based on EEG spectrum', Communications in Computer and Information Science, pp. 450-454.
View/Download from: Publisher's site
Distraction while driving is a serious problem that can have many catastrophic consequences. Developing a countermeasure to detect the drivers' distraction is imperative. This study measured Electroencephalography (EEG) signals from six healthy participants while they were asked to pay their full attention to a lane-keeping driving task or a math problem-solving task. The time courses of six distinct brain networks (Frontal, Central, Parietal, Occipital, Left Motor, and Right Motor) separated by Independent Component Analysis were used to build the distraction-detection model. EEG data were segmented into 400-ms epochs. Across subjects, 80% of the EEG epochs were used to train various classifiers that were tested against the remaining 20% of the data. The classification performance based on support vector machines (SVM) with a radial basis function (RBF) kernel achieved accuracy of 84.7±2.7% or 85.8±1.3% for detecting subjects' focuses of attention to the math-solving or lane-deviation task, respectively. The high attention-detection accuracy demonstrated the feasibility of accurately detecting drivers' attention based on the brain activities. This demonstration may lead to a practical real-time distraction-detection system for improving road safety. © Springer-Verlag Berlin Heidelberg 2013.
Huang, C.S., Lin, C.L., Yang, W.Y., Ko, L.W., Liu, S.Y. & Lin, C.T. 2013, 'Applying the fuzzy C-means based dimension reduction to improve the sleep classification system', IEEE International Conference on Fuzzy Systems.
View/Download from: Publisher's site
Having a well sleep quality is important factor in our daily life. The evaluation of sleep stages has become an important issue due to the distribution of sleep stages across a whole night relates to sleep quality. This study aims to propose a sleep classification system, consists of a preliminary wake detection rule, sleep feature extraction, fuzzy c-means based dimension reduction, support vector machine with radial basis function kernel, and adaptive adjustment scheme, with only FP1 and FP2 electroencephalography. Compared with the results from the sleep technologist, the average accuracy and Kappa coefficient of the proposed sleep classification system is 70.92% and 0.6130, respectively, for individual 10 normal subjects. Thus, the proposed sleep classification system could provide a preliminary report of sleep stages to assistant doctors to make decision if a patient needs to have a detailed testing in a sleep laboratory. © 2013 IEEE.
Li, S.Y., Ko, L.W., Lin, C.T., Tam, L.M., Chen, H.K. & Lao, S.K. 2013, 'System modeling and synchronization of nonlinear chaotic systems with uncertainty and disturbance by innovative fuzzy modeling strategy', IEEE International Conference on Fuzzy Systems.
View/Download from: Publisher's site
In this paper, an application of the innovative fuzzy model [1] is applied to simulate and synchronize two classical Sprott chaotic systems with unknown noise and disturbance. In traditional Takagi-Sugeno fuzzy (T-S fuzzy) model, there will be 2N linear subsystems (according to 2N fuzzy rules) and m 2N equations in the T-S fuzzy system, where N is the number of minimum nonlinear terms and m is the order of the system. Through the new fuzzy model, a complicated nonlinear system is linearized to a simple form - linear coupling of only two linear subsystems and the numbers of fuzzy rules can be reduced from 2N to 2 N. The fuzzy equations become much simpler. There are two Sprott systems in numerical simulations to show the effectiveness and feasibility of new model. © 2013 IEEE.
Wu, S.L., Wu, C.W., Pal, N.R., Chen, C.Y., Chen, S.A. & Lin, C.T. 2013, 'Common spatial pattern and linear discriminant analysis for motor imagery classification', Proceedings of the 2013 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain, CCMB 2013 - 2013 IEEE Symposium Series on Computational Intelligence, SSCI 2013, pp. 146-151.
View/Download from: Publisher's site
A Brain-Computer Interface (BCI) system provides a convenient way of communication for healthy subjects and subjects who suffer from severe diseases such as amyotrophic lateral sclerosis (ALS). Motor imagery (MI) is one of the popular ways of designing BCI systems. The architecture of many BCI system is quite complex and they involve time consuming processing. The electroencephalography (EEG) signal is the most commonly used inputs for BCI applications but EEG is often contaminated with noise. To overcome such drawbacks, in this paper we use the common spatial pattern (CSP) for feature extraction from EEG and the linear discriminant analysis (LDA) for motor imagery classification. In this study, CSP and LDA have been used to reduce the artifact and classify MI-based EEG signal. We have used two-level cross validation scheme to determine the subject specific best time window and number of CSP features. We have compared the performance of our system with BCI competition results. We have also experimented with MI data generated in our lab. The proposed system is found to produce good results. In particular, using our EEG data for MI movements, we have obtained an average classification accuracy of 80% for two subjects using only 9 channels, without any feature selection. This proposed MI-based BCI system may be used in real life applications. © 2013 IEEE.
Huang, C.S., Lin, C.L., Ko, L.W., Liu, S.Y., Sua, T.P. & Lin, C.T. 2013, 'A hierarchical classification system for sleep stage scoring via forehead EEG signals', Proceedings of the 2013 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain, CCMB 2013 - 2013 IEEE Symposium Series on Computational Intelligence, SSCI 2013, pp. 1-5.
View/Download from: Publisher's site
The study adopts the structure of hierarchical classification to develop an automatic sleep stage classification system using forehead (Fpl and Fp2) EEG signals. The hierarchical classification consists of a preliminary wake detection rule, a novel feature extraction method based on American Academy of Sleep Medicine (AASM) scoring manual, feature selection methods and SVM. After estimating the preliminary sleep stages, two adaptive adjustment schemes are applied to adjust the preliminary sleep stages and provide the final estimation of sleep stages. Clinical testing reveals that the proposed automatic sleep stage classification system is about 77% accuracy and 67% kappa for individual 10 normal subjects. This system could provide the possibility of long term sleep monitoring at home and provide a preliminary result of sleep stages so that doctor could decide if a patient needs to have a detailed diagnosis using Polysomnography (PSG) system in a sleep laboratory of hospital. © 2013 IEEE.
Lin, C.T., Wang, Y.K., Fan, J.W. & Chen, S.A. 2013, 'The influence of acute stress on brain dynamics', Proceedings of the 2013 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain, CCMB 2013 - 2013 IEEE Symposium Series on Computational Intelligence, SSCI 2013, pp. 6-10.
View/Download from: Publisher's site
Living under high stress may be unhealthy. This study explores electroencephalography (EEG) correlated with stressful circumstances by using the task-switching paradigm with feedback information. According to the behavioral and physiological evidence, acute stress created by this paradigm affected the performance of participants. Under stress, the participants responded quickly and inaccurately. The EEG results correlated with acute stress were found in the frontal midline cortex, especially on the theta and alpha bands. These specific factors are considered importance features for detecting the influence of stress by applying various machine-learning methods and neuro-fuzzy systems. This comprehensive study can provide knowledge for studying stress and designing Brain-Computer Interface (BCI) systems in the future. © 2013 IEEE.
Ko, L.W., Lee, H.C., Tsai, S.F., Shih, T.C., Chuang, Y.T., Huang, H.L., Ho, S.Y. & Lin, C.T. 2013, 'EEG-based motion sickness classification system with genetic feature selection', Proceedings of the 2013 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain, CCMB 2013 - 2013 IEEE Symposium Series on Computational Intelligence, SSCI 2013, pp. 158-164.
View/Download from: Publisher's site
People tend to get motion sickness on a moving boat, train, airplane, car, or amusement park rides. Many previous studies indicated that motion sickness sometimes led to traffic accidents, so it becomes an important issue in our daily life. In this study, we designed a VR-based motion-sickness platform with a 32-channel EEG system and a joystick which is used to report the motion sickness level (MSL) in real time during experiments. The results show it is feasible to estimate subject's MSL based on re-sampling frequency band proved by the high test accuracy. A comparison between general prediction models (such as LDA, QDA, KNN) and IBCGA shows that the IBCGA can be effectively increase the accuracy. In this paper, an extended-IBCGA (e-IBCGA) is proposed and it provides more accuracy than the prior-art research. The test results show that e-IBCGA increases at least 10% to 20% test accuracy in 6 subjects. © 2013 IEEE.
Chuang, C.H., Lin, Y.P., Ko, L.W., Jung, T.P. & Lin, C.T. 2013, 'Automatic design for independent component analysis based brain-computer interfacing', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 2180-2183.
View/Download from: Publisher's site
This study proposes a new framework, independent component ensemble, to leverage the acquired knowledge into a truly automatic and on-line EEG-based brain-computer interfacing (BCI). The envisioned design includes: (1) independent source recover using independent component analysis (ICA) (2) automatic selection of the independent components of interest (ICi) associated with human behaviors; (3) multiple classifiers with a parallel constructing and processing structure; and (4) a simple fusion scheme to combine the decisions from multiple classifiers. Its implications in BCI are demonstrated through a sample application: cognitive-state monitoring of participants performing a realistic sustained-attention driving task. Empirical results showed the proposed ensemble design could provide an improvement of 7%15% in overall accuracy for the classification of the arousal state and the driving performance. In summary, constructing ICi-ensemble classifiers and combining their outputs demonstrates a practical option for ICA-based BCIs to reduce the risk of not obtaining any desired independent source or selecting an inadequate component. Most importantly, the ensemble design for integrating information across multiple brain areas creates potentials for developing more complicated BCIs for real world applications. © 2013 IEEE.
Lin, Y.Y., Chang, J.Y. & Lin, C.T. 2013, 'An interval type-2 neural fuzzy inference system (IT2NFIS) with compensatory operator', Proceedings of the 2013 Joint IFSA World Congress and NAFIPS Annual Meeting, IFSA/NAFIPS 2013, pp. 884-889.
View/Download from: Publisher's site
In this paper, an interval type-2 neural fuzzy system (IT2NFIS) with compensatory operator is proposed for system modeling. The IT2NFIS uses type-2 fuzzy sets in the premise clause in order to effectively handle the uncertainties in terms of data and information. The premise part of each compensatory fuzzy rule is an interval type-2 fuzzy set in the IT2NFIS, where compensatory operation is able to adaptively adjust fuzzy membership functions and to dynamically optimize fuzzy operations. The consequent part in the IT2NFIS consists of the Takagi-Sugeno-Kang (TSK) type that is a linear combination of exogenous input variables. Initially the rule base in the IT2NFIS is empty. All rules generated are based on on-line type-2 fuzzy clustering. All free weights are learned by a gradient descent algorithm to improve the learning performance. Simulation results show that our approach yields smaller root mean squared errors than its rivals. © 2013 IEEE.
Prasad, M., Lin, C.T., Yang, C.T. & Saxena, A. 2013, 'Vertical collaborative fuzzy C-means for multiple EEG data sets', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 246-257.
View/Download from: Publisher's site
Vertical Collaborative Fuzzy C-Means (VC-FCM) is a clustering method that performs clustering on a data set of having some set of patterns with the collaboration of some knowledge which is obtained from other data set having the same number of features but different set of patterns. Uncertain relationship lies in data between the data sets as well as within a dataset. Practically data of the same group of objects are usually stored in different datasets; in each data set, the data dimensions are not necessarily the same and unreal data may exist. Fuzzy clustering of a single data set would bring about less reliable results. And these data sets cannot be integrated for some reasons. An interesting application of vertical clustering occurs when dealing with huge data sets. Instead of clustering them in a single pass, we split them into individual data sets, cluster each of them separately, and reconcile the results through the collaborative exchange of prototypes. Vertical collaborative fuzzy C-Means is a useful tool for dealing collaborative clustering problems where a feature space is described in different pattern-sets. In this paper we use collaborative fuzzy clustering, first we cluster each data set individually and then optimize in accordance with the dependency of these datasets is adopted so as to improve the quality of fuzzy clustering of a single data set with the help of other data sets, taking personal privacy and security of data into consideration. © 2013 Springer-Verlag Berlin Heidelberg.
Kellihan, B., Doty, T.J., Hairston, W.D., Canady, J., Whitaker, K.W., Lin, C.T., Jung, T.P. & McDowell, K. 2013, 'A real-world neuroimaging system to evaluate stress', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 316-325.
View/Download from: Publisher's site
While the laboratory setting offers researchers a great deal of experimental control, this environment also limits how generalizable the results are to the real world. This is particularly true when studying the multifaceted phenomenon of stress, which often relies on personal experience, a dimension that is difficult to reproduce in the laboratory setting. This paper describes a novel, multi-aspect real-world integrated neuroimaging system (MARIN) optimized to study physiological phenomena in the real-world and particularly suited to the study of stress. This system integrates neurological data from a gel-free, wireless EEG device with physiological data from wireless cardiac and skin conductance sensors, as well as self-reports of activity and stress. Coordination of the system is managed through an Android handheld mobile device that also logs salient events and presents inventories for subjective reports of stress. The integration of these components creates a rich, multimodal dataset with minimal interference to the user's daily life, and these data will guide the further understanding of neurological mechanisms of stress. © 2013 Springer-Verlag Berlin Heidelberg.
Ko, L.W., Chuang, C.H., Huang, C.S., Chen, Y.H., Lu, S.W., Liao, L.D., Chang, W.T. & Lin, C.T. 2013, 'Real-time vigilance estimation using mobile wireless mindo EEG device with spring-loaded sensors', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 450-458.
View/Download from: Publisher's site
Monitoring the neurophysiological activities of human brain dynamics in an operational environment poses a severe measurement challenge using current laboratory-oriented biosensor technology. The goal of this research is to design, develop and test the wearable and wireless dry-electrode EEG human-computer interface (HCI) that can allow assessment of brain activities of participants actively performing ordinary tasks in natural body positions and situations within a real operational environment. Its implications in HCI were demonstrated through a sample application: vigilance-state prediction of participants performing a realistic sustained-attention driving task. Besides, this study further developed an online signal processing for extracting EEG features and assessing cognitive performance. We demonstrated the feasibility of using dry EEG sensors and miniaturized supporting hardware/software to continuously collect EEG data recorded from hairy sites (i.e., occipital region) in a realistic VR-based dynamic driving simulator. © 2013 Springer-Verlag Berlin Heidelberg.
Lin, C.T., Chuang, C.H., Huang, C.S., Chen, Y.H. & Ko, L.W. 2013, 'Real-time assessment of vigilance level using an innovative Mindo4 wireless EEG system', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 1528-1531.
View/Download from: Publisher's site
Monitoring the neurophysiological activities of driver in an operational environment poses a severe measurement challenge using a current laboratory-oriented biosensor technology. The aims of this research are to 1) introduce a dry and wireless EEG system used for conveniently recording EEG signals from forehead regions, 2) propose an effective system for processing EEG recordings and translating them into the vigilance level, and 3) implement the proposed system with a JAVA-based graphical user interface (GUI) for online analysis. To validate the performance of the proposed system, this study recruited eight voluntary subjects to participate a 90-min sustained-attention driving task in a virtual-realistic driving environment. Physiological evidence obtained from the power spectral analysis showed that the dry EEG system could distinguish an alert EEG from a drowsy EEG by evaluating the spectral dynamics of delta and alpha activities. Furthermore, the experimental result of the comparison of the prediction performance using four forehead electrode sites (AF8, FP2, FP1, and AF7) implied that a single-electrode EEG signal used in the mobile and wireless EEG system is able to obtain a high prediction accuracy (93%). Taken together, the proposed system applied a dry-EEG device combined with an effective algorithm can be a promising technology for real driving applications. © 2013 IEEE.
Liao, L.D., Chen, Y.Y., Lin, C.T. & Li, M.L. 2013, 'Functional photoacoustic micro-imaging of cerebral hemodynamic changes in single blood vessels after photo-induced brain stroke', Progress in Biomedical Optics and Imaging - Proceedings of SPIE.
View/Download from: Publisher's site
Studying the functional hemodynamic roles of individual cerebral cortical arterioles in maintaining both the structure and function of cortical regions during and after brain stroke in small animals is an important issue. Recently, functional photoacoustic microscopy (fPAM) has been proved as a reliable imaging technique to probe the total hemoglobin concentration (HbT), cerebral blood volume (CBV) and hemoglobin oxygen saturation (SO2) in single cerebral blood vessels of rats. Here, we report the application of fPAM associated with electrophysiology recordings to investigating functional hemodynamic changes in single cortical arterioles of rats with electrical forepaw stimulation after photo-induced ischemic stroke. Because of the weak optical focusing nature of our fPAM system, photo-induced ischemic stroke targeting single cortical arterioles can be easily conducted with simple adaptation. Functional HbT, CBV and SO2 changes associated with the induced stroke in selected arterioles from the anterior cerebral artery system were imaged with 36 x 65-m spatial resolution. Experimental results showed that after photo-occlusion of a single arteriole, the functional changes of nearby arterioles in cerebral cortex only can be observed immediately after the stroke. After a few minutes of stroke onset, there are no significant functional changes under the forepaw stimulation, suggesting that alternate blood flow routes are not actively recruited. The fPAM with electrophysiology recordings complements existing imaging techniques and has the potential to offer a favorable tool for explicitly studying cerebral hemodynamics in small animal models of photo-indcued ischemic stroke. © 2013 Copyright SPIE.
Wang, Y.T., Cheng, C.K., Huang, K.C., Lin, C.T., Wang, Y. & Jung, T.P. 2012, 'Cell-phone based Drowsiness Monitoring and Management system', 2012 IEEE Biomedical Circuits and Systems Conference: Intelligent Biomedical Electronics and Systems for Better Life and Better Environment, BioCAS 2012 - Conference Publications, pp. 200-203.
View/Download from: Publisher's site
In recent years, driving fatigue and cognitive lapses have gained increasing attentions in the fields of public security, especially regarding the safe manipulation of vehicles. Studies have explored the relation between electroencephalogram (EEG) and driving behavior and several of them further proposed effective approaches (e.g., auditory feedback) to arouse subjects from drowsiness. However, these studies were performed under laboratory-oriented configurations using tethered, ponderous EEG equipment. In real environments, the setup of bulky cognitive monitoring equipment with a long-prep time is not feasible. Therefore, this study extends previous laboratory work by developing an on-line Drowsiness Monitoring and Management (DMM) System featuring a mobile wireless dry-sensor EEG headgear and a cell-phone based real-time EEG processing platform. The DMM system can continuously observe EEG dynamics, deliver arousing feedback to users experiencing momentary cognitive lapses, and assess the efficacy of the feedback in near real-time. © 2012 IEEE.
Chang, J.Y., Liao, S.H. & Lin, C.T. 2012, 'Adaptive least trimmed squares fuzzy neural network', 2012 International Conference on Fuzzy Theory and Its Applications, iFUZZY 2012, pp. 413-416.
View/Download from: Publisher's site
In this paper, we propose the adaptive least trimmed squares fuzzy neural network (ALTS-FNN), which applies the scale estimate to the least trimmed squares fuzzy neural network (LTS-FNN). The emphasis of this paper is particular on the robustness against the outliers and the choice of the trimming constant can be determined adaptively. Some numerical examples will be provided to compare the robustness against outliers for usual FNN and the ALTS-FNN. Simulation results show that the ALTS-FNN in the paper have good performance for outlier detection. © 2012 IEEE.
Chen, S.H., Lin, C.T. & Fu, L.C. 2012, 'Second order sliding mode control on task-space of a 6-DOF Stewart platform', IECON Proceedings (Industrial Electronics Conference), pp. 2482-2487.
View/Download from: Publisher's site
In this paper, the second order sliding mode control strategy is applied on a 6-dof parallel manipulator, so-called Stewart platform. The advantages such as robustness and design simplicity of sliding mode control are the main reason that it was often used on robotic systems. However, the chattering phenomenon of conventional sliding mode control may cause unstability due to the practical implementations of the switching control. As a result of discontinuous control signal, the chattering phenomenon can be eliminated via designing a higher order control strategy. Once the switching signal is designed in higher order manifold, the real input will be smoothed because of the integration. The paper represents a robustness control method without chattering for Stewart platform which is a multi-input nonlinear system. The control performance of the second-order approach was compared to conventional sliding mode control in the simulation results. © 2012 IEEE.
Wu, S.L., Liao, L.D., Liou, C.H., Chen, S.A., Ko, L.W., Chen, B.W., Wang, P.S., Chen, S.F. & Lin, C.T. 2012, 'Design of the multi-channel electroencephalography-based brain-computer interface with novel dry sensors', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 1793-1797.
View/Download from: Publisher's site
The traditional brain-computer interface (BCI) system measures the electroencephalography (EEG) signals by the wet sensors with the conductive gel and skin preparation processes. To overcome the limitations of traditional BCI system with conventional wet sensors, a wireless and wearable multi-channel EEG-based BCI system is proposed in this study, including the wireless EEG data acquisition device, dry spring-loaded sensors, a size-adjustable soft cap. The dry spring-loaded sensors are made of metal conductors, which can measure the EEG signals without skin preparation and conductive gel. In addition, the proposed system provides a size-adjustable soft cap that can be used to fit user's head properly. Indeed, the results are shown that the proposed system can properly and effectively measure the EEG signals with the developed cap and sensors, even under movement. In words, the developed wireless and wearable BCI system is able to be used in cognitive neuroscience applications. © 2012 IEEE.
Huang, K.-.C., Jung, T.-.P., Chuang, C.-.H., Ko, L.-.W. & Lin, C.-.T. 2012, 'Preventing lapse in performance using a drowsiness monitoring and management system.', Conf Proc IEEE Eng Med Biol Soc, United States, pp. 3336-3339.
View/Download from: Publisher's site
Research on public security, especially the safe manipulation and control of vehicles, has gained increasing attention in recent years. This study proposes a closed-loop drowsiness monitoring and management system that can estimate subjects' driving performance. The system observes electroencephalographic (EEG) dynamics and behavioral changes, delivers arousing feedback to individuals experiencing momentary cognitive lapses, and assesses the efficacy of the feedback. Results of this study showed that the arousing feedback immediately improved subject performance, which was accompanied by concurrent theta- and alpha-power suppression in the bilateral occipital areas. This study further demonstrated the feasibility of accurately assessing the efficacy of arousing feedback presented to drowsy participants by monitoring the changes in their EEG power spectra.
Lin, C.L., Jung, M., Wu, Y.C., Lin, C.T. & She, H.C. 2012, 'Brain dynamics of mathematical problem solving', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 4768-4771.
View/Download from: Publisher's site
The purpose of this study is to examine brain activities of participants solving mental math problems. The research investigated how problem difficulty affected the subjects' responses and electroencephalogram (EEG) in different brain regions. In general, it was found that solution latencies (SL) to the math problems increased with difficulty. The EEG results showed that across subjects, the right-central beta, left-parietal theta, left-occipital theta and alpha, right-parietal alpha and beta, medial-frontal beta and medial central theta power decreased as task difficulty increased. This study further explored the effects of problem-solving performance on the EEG. Slow solvers exhibited greater frontal theta activities in the right hemisphere, whereas an inverse pattern of hemispheric asymmetry was found in fast solvers. Furthermore, analyses of spatio-temporal brain dynamics during problem solving show progressively stronger alpha- and beta-power suppression and theta-power augmentation as subjects were reaching a solution. These findings provide a better understanding of cortical activities mediating math-based problem solving and knowledge acquisition that can ultimately benefit math learning and education. © 2012 IEEE.
Chang, J.Y., Han, M.F. & Lin, C.T. 2012, 'Optimization of fuzzy systems using group-based evolutionary algorithm', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 291-298.
View/Download from: Publisher's site
This paper proposes a group-based evolutionary algorithm (GEA) for the fuzzy system (FS) optimization. Initially, we adopt an entropy measure method to determine the number of rules. Fuzzy rules are automatically generated from training data by entropy measure. Subsequently, the GEA is performed to optimize all the free parameters for the FS design. In the evolution process, a FS is coded as an individual. All individuals based on their performance are partitioned into a superior group and an inferior group. The superior group, which is composed of individuals with better performance, uses a global evolution operation to search potential individuals. In the inferior group, individuals with a worse performance employ the local evolution operation to search better individuals near the current best individual. Finally, the proposed FS with GEA model (FS-GEA) is applied to time series forecasting problem. Results show that the proposed FS-GEA model obtains better performance than other algorithm. © 2012 Springer-Verlag.
Tu, T.Y., Chao, P.C.P. & Lin, C.T. 2012, 'A new liquid crystal lens with axis-tunability via three sector electrodes', Microsystem Technologies, pp. 1297-1307.
View/Download from: Publisher's site
A novel liquid crystal (LC) lens with an on-line tunability on focus length and optical axis is proposed in this study. The designed lens has a LC layer sandwiched by two ITO glasses, one of which is patterned with three sector electrodes. With varied sets of pre-designed voltages applied to these three electrodes, the LC lens can not only render focusing effects but also tunability on the optical axis of the lens to an arbitrary axis. A vector-form equation is developed to predict the direction of axis tuning. Simulations are next conducted to predict dynamics of the LCs in the lens and also the focusing and axis-tuning properties of the lens. Important sizes and materials and fabrication process of the lens are determined and optimized based on simulation results. The designed LC lens is fabricated, and then experiments are conducted to demonstrate the performance of the designed LC lens on axis tuning. It shows that the focusing axis of the LC lens can be effectively changed by pre-calculated combinations of three voltages. It is also shown that the average movement of the focal point per applied voltage reaches 4.778 lm/V. © Springer-Verlag 2012.
Ou, C.Z., Lin, B.S., Chang, C.J. & Lin, C.T. 2012, 'Brain computer interface-based smart environmental control system', Proceedings of the 2012 8th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2012, pp. 281-284.
View/Download from: Publisher's site
A Brain computer interface-based Smart Environmental Control System (BSECS) was proposed in this study. Many environmental control systems have been proposed to improve human life quality recently. However, few researches focused on environment control by using human's physiological state directly. Based on the advantage of our technique on Brain Computer Interface (BCI), we combined BCI technique with Universal Plug and Play (UPnP) home networking for smart house applications. Our BSECS mainly consists of a wireless physiological signal acquisition module, an embedded signal processing module, a Simple Control Protocol (SCP)/ Power Line Communication (PLC) environment controller, and a host system. Here, the physiological signal acquisition module and embedded signal processing module were designed for long-term electroencephalogram (EEG) monitoring and real-time cognitive state detection respectively. The advantages of low-power consumption and small volume of the above modules are suitable for smart house applications in daily life. The proposed BSECS has been verified in a practical demo room, and can be simply extended and integrated with the UPnP home networking for other applications. © 2012 IEEE.
Khoo, I.H., Reddy, H.C., Van, L.D. & Lin, C.T. 2012, 'Delta operator based 2-D VLSI filter structures without global broadcast and incorporation of the quadrantal symmetry', ISCAS 2012 - 2012 IEEE International Symposium on Circuits and Systems, pp. 3190-3193.
View/Download from: Publisher's site
Having local data communication (without global broadcast of signals) among the elements is important in VLSI designs. Recently, 2-D systolic digital filter architectures were presented which eliminated the global broadcast of the input and output signals. The delta discrete-time operator based 1-D and 2-D digital filters (in -domain) were shown to offer better numerical accuracy and lower coefficient sensitivity in narrowband filter designs when compared to the traditional shiftoperator formulation. Further, the complexity in the design and implementation of 2-D filters can be reduced considerably if the symmetries that might be present in the frequency responses of these filters are utilized. With this motivation we present new 2-D VLSI filter structures, without global broadcast, using delta discrete-time operator for the first time. We also present frame works in -domain that realizes 2-D filters possessing quadrantal symmetry in its magnitude response. The separable denominator and quadrantal symmetry structures have the advantage of reduced number of multipliers while ensuring the 2-D filter stability. © 2012 IEEE.
Yang, S.R., Hsu, S.C., Lu, S.W., Ko, L.W. & Lin, C.T. 2012, 'Development of adaptive QRS detection rules based on center differentiation method for clinical application', ISCAS 2012 - 2012 IEEE International Symposium on Circuits and Systems, pp. 2071-2074.
View/Download from: Publisher's site
Interpretation of cardiac rhythms is highly dependent on accurate detection of atrial activity. The robustness is an important requirement for clinical usage. This study presents an adaptive QRS detection method for real-time clinical ECG signals. In this method, center differentiation is applied as a whitening filer, and a composite function enhances the high frequency QRS energy. To robustly detect clinical data, a novel threshold selection method based on statistics is proposed. Moreover, this study provides a benchmarking clinical dataset acquired from cardiac patients. Our extensive experimental results using the MIT-BIH arrhythmia database show that our technique can detect beats with 99.67% accuracy, and the sensitivity is 99.83%. With the exceptional QRS detection result, further testing of the proposed method with clinical data shows the accuracy for atrial and ventricular arrhythmias is 82.9% and 90.2%, respectively. © 2012 IEEE.
Yang, S.R., Shi-An, C., Shu-Fang, T. & Lin, C.T. 2012, 'Transcutaneous electrical nerve stimulation system for improvement of flight orientation in a VR-based motion environment', ISCAS 2012 - 2012 IEEE International Symposium on Circuits and Systems, pp. 2055-2058.
View/Download from: Publisher's site
It is very important to reduce the possibility of spatial disorientation because spatial disorientation is a major cause of aircraft crashes. In this research, we assess the effect of transcutaneous electrical nerve stimulation (TENS) on spatial cognitive function by measuring physiological signals, including brain waves measured by electroencephalography (EEG). Through physiological signals such as brain waves, we can objectively understand a subject's cognitive state. Moreover, we use virtual reality technology to build 3D scenery that provides navigation clues in an environment with no visual reference frame. Using the virtual environment, we confirm that TENS lessens spatial disorientation. Across subjects and sessions, the parietal component of brain activity exhibited baseline elevation predominantly in the theta (4-7 Hz) and alpha (8-12 Hz) band as the subjects navigated. This study also found that subjects performed better after TENS in terms of behavioral and EEG data. The results facilitate understanding of the cognitive function of maintaining spatial orientation and development of a device for assisting pilots and reducing the occurrence of spatial disorientation. © 2012 IEEE.
Lin, C.T., Tsai, S.F., Lee, H.C., Huang, H.L., Ho, S.Y. & Ko, L.W. 2012, 'Motion sickness estimation system', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: Publisher's site
Motion sickness occurs when the brain receives conflicting sensory information from body, inner ear and eyes [1]. In some cases, a decreased ability to actively control the body's postural motion also causes motion sickness [2][3]. Many previous studies have indicated that motion sickness had negative effect on driving performance, and sometimes lead to serious traffic accidents due to self-control ability decline. Therefore motion sickness becomes a very important issue in our daily life especially considering driving safety. There are many attempts made by researchers to realize motion sickness, and detect motion sickness in the early stage. Although many motion-sickness-related biomarkers have been identified, estimating human motion sickness level (MSL) remains a challenge in operational environment. In our past studies, we found that features in the occipital area were highly correlated with the driver's driving performance. In this study, we designed a virtual-reality (VR) based driving environment with instinct-MSL-reporting mechanism. When a subject performed a driving task, his/her brain EEG was recorded simultaneously. From those EEG data, features associated with left motor brain area, parietal brain area and occipital midline brain area which predicted MSL were extracted by an optimal classifier implemented by an inheritable bi-objective combinatorial genetic algorithm (IBCGA) with support vector machine. Unlike traditional correlation-based method, IBCGA aims to select a small set of EEG features and maximize the prediction accuracy simultaneously in BCI applications. Once the optimal feature set predicting MSL is successfully found, a driver's cognitive state can be monitored. © 2012 IEEE.
Lin, C.T., Wang, Y.K. & Chen, S.A. 2012, 'A hierarchal classifier for identifying independent components', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: Publisher's site
Brain-computer interface (BCI) has shown explosive growth for multiple applications in the recently years. Removing artifacts and selecting useful brain sources are essential in BCI research. Independent Component Analysis (ICA) has been proven as an effective technique to remove artifacts and many brain related researches are based on ICA. However, the useful independent components with brain sources are usually selected manually according to the scalp-plots. This is great inconvenience and a barrier for real-time BCI applications of EEG. In this investigation, a two-layer automatic identification model is proposed to select useful brain sources. It is based on neural network including support vector machine with radial basis function (SVMRBF) and self-organizing map (SOM). In the first layer, SVM discriminates useful independent components from the artifact effectively. In the second layer, these selected useful components are automatically classified to different spatial brain sources according to SOM. This study suggests this model to one general application for EEG study. It can reduce the effect of subjective judgment and improve the performance of EEG analysis. © 2012 IEEE.
Chuang, C.H., Huang, C.S., Lin, C.T., Ko, L.W., Chang, J.Y. & Yang, J.M. 2012, 'Mapping information flow of independent source to predict conscious level: A granger causality based brain-computer interface', Proceedings - 2012 International Symposium on Computer, Consumer and Control, IS3C 2012, pp. 813-816.
View/Download from: Publisher's site
Recent studies have shown that the various brain networks over different cognitive states. In contrast to measure a physiological change over a single region, the information flows between brain regions described by effective connectivity provides an informative dynamic over the whole brain. In this study, we proposed a source information flow network based on the combination of Granger causality and support vector regression to predict driver's conscious level. This work provides the first application of using brain network to develop a brain-computer interface and obtain a sound result of performance. © 2012 IEEE.
Liao, L.D., Lin, C.T., McDowell, K., Wickenden, A.E., Gramann, K., Jung, T.P., Ko, L.W. & Chang, J.Y. 2012, 'Biosensor technologies for augmented brain-computer interfaces in the next decades', Proceedings of the IEEE, pp. 1553-1566.
View/Download from: Publisher's site
The study of brain-computer interfaces (BCIs) has undergone 30 years of intense development and has grown into a rich and diverse field. BCIs are technologies that enable direct communication between the human brain and external devices. Conventionally, wet electrodes have been employed to obtain unprecedented sensitivity to high-temporal-resolution brain activity; recently, the growing availability of various sensors that can be used to detect high-quality brain signals in a wide range of clinical and everyday environments is being exploited. This development of biosensing neurotechnologies and the desire to implement them in real-world applications have led to the opportunity to develop augmented BCIs (ABCIs) in the upcoming decades. An ABCI is similar to a BCI in that it relies on biosensors that record signals from the brain in everyday environments; the signals are then processed in real time to monitor the behavior of the human. To use an ABCI as a mobile brain imaging technique for everyday, real-life applications, the sensors and the corresponding device must be lightweight and the equipment response time must be short. This study presents an overview of the wide range of biosensor approaches currently being applied to ABCIs, from their use in the laboratory to their application in clinical and everyday use. The basic principles of each technique are described along with examples of current applications of cutting-edge neuroscience research. In summary, we show that ABCI techniques continue to grow and evolve, incorporating new technologies and advances to address ever more complex and important neuroscience issues, with advancements that are envisioned to lead to a wide range of real-life applications. © 2012 IEEE.
Lin, C.T. & McDowell, K. 2012, 'Prolog to the section on neurotechnological systems: The brain-computer interface', Proceedings of the IEEE, pp. 1551-1552.
View/Download from: Publisher's site
Current and potential brain-computer interface (BCI) technologies and research enabled by recent advances in wearable, mobile biosensors and data acquisition, neuroscience, computational and analytical approaches, and computing for brain imaging in real-world environments, is presented. Liao et al. discuss barriers to taking brain imaging systems out of laboratory and clinical settings and into everyday environments, and highlight current and future approaches to address those barriers. Makeig et al. discuss the challenges associated with building robust and useful BCI models from accumulated biological knowledge and available data, and the technical problems associated with incorporating multimodal physiological, behavioral, and contextual data that may become ubiquitous in the future. This paper discusses past and current BCI applications and proposes future BCI technologies that will make significant expansion into training, education, entertainment, rehabilitation, and human-system performance domains.
Liao, L.D., Chen, Y.Y., Lin, C.T., Chang, J.Y. & Li, M.L. 2012, 'Functional photoacoustic micro-imaging of rat cerebral hemodynamic response function in single vessels during forepaw electrical stimulation', Progress in Biomedical Optics and Imaging - Proceedings of SPIE.
View/Download from: Publisher's site
The specificity of the hemodynamic response function (HRF) is determined spatially by the vascular architecture and temporally by the evolution of hemodynamic changes. Here, we used functional photoacoustic microscopy (fPAM) to investigate the spatiotemporal evolution of the HRFs of hemoglobin concentration (HbT), cerebral blood volume (CBV) and hemoglobin oxygen saturation (SO 2) in single cerebral vessels to rat left-forepaw stimulation. The HRF changes in specific cerebral vessels responding to different stimulation intensities and durations were bilaterally imaged with 36 65-m spatial resolution. Various electrical stimulations were applied with stimulation intensities at 1, 2, 6 and 10-mA combined with 5-s and 15-s stimulation durations, respectively. Our main findings were as follows: 1) the functional HbT and SO 2 increased sub-linearly with increasing stimulus intensities and 2) the results suggested that the CBV changes are more linearly correlated with arterioles than HbT and SO 2 within a limited dynamic range of stimulation intensities and duration. The findings in this study indicate that the regulation of hemodynamic changes in single cerebral vessels can be reliable studied by the fPAM technique without the use of contrast agents. © 2012 SPIE.
Lin, Y.P., Chen, J.H., Duann, J.R., Lin, C.T. & Jung, T.P. 2011, 'Generalizations of the subject-independent feature set for music-induced emotion recognition', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 6092-6095.
View/Download from: Publisher's site
Electroencephalogram (EEG)-based emotion recognition has been an intensely growing field. Yet, how to achieve acceptable accuracy on a practical system with as fewer electrodes as possible is less concerned. This study evaluates a set of subject-independent features, based on differential power asymmetry of symmetric electrode pairs [1], with emphasis on its applicability to subject variability in music-induced emotion classification problem. Results of this study have evidently validated the feasibility of using subject-independent EEG features to classify four emotional states with acceptable accuracy in second-scale temporal resolution. These features could be generalized across subjects to detect emotion induced by music excerpts not limited to the music database that was used to derive the emotion-specific features. © 2011 IEEE.
Wei, C.S., Chuang, S.W., Wang, W.R., Ko, L.W., Jung, T.P. & Lin, C.T. 2011, 'Implementation of a motion sickness evaluation system based on EEG spectrum analysis', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 1081-1084.
View/Download from: Publisher's site
Motion sickness is a normal response to real, perceived, or even anticipated movement. People tend to get motion sickness on a moving boat, train, airplane, car, or amusement park rides. Motion sickness occurs when the body, the inner ear, and the eyes send conflicting signals to the brain. Sensory conflict theory that came about in the 1970's has become the most widely accepted theorem of motion-sickness among scientists [1]. The theory proposed that the conflict between the incoming sensory inputs could induce motion-sickness. However, some new research studies have appeared to tackle the issue of the vestibular function in central nervous system (CNS). In the previous human subject studies, researchers attempt to confirm the brain areas involved in the conflict in multi-modal sensory systems by means of clinical or anatomical methods. Our past studies had investigated the EEG activities correlated with motion sickness in a virtual-reality based driving simulator. We found that the parietal, motor, occipital brain regions exhibited significant EEG power changes in response to vestibular and visual stimuli. Based on these experimental results, we attempt to implement an EEG-based evaluation system to estimate subject's motion sickness level upon the major EEG power spectra from these motion sickness related brain area in this study. The evaluation system can be applied to early detect the subject's motion sickness level and prevent the uncomfortable syndromes occurred in advance in our daily life. © 2011 IEEE.
Chen, C.H., Lin, C.J. & Lin, C.T. 2011, 'An immune symbiotic evolution learning for compensatory neural fuzzy networks and its applications', IEEE International Conference on Fuzzy Systems, pp. 2819-2826.
View/Download from: Publisher's site
This study presents an efficient immune symbiotic evolution learning algorithm for the compensatory neural fuzzy network (CNFN). The proposed immune symbiotic evolution learning method (ISEL) includes three major components initial population, subgroup symbiotic evolution and immune system algorithm. The advantage of the proposed ISEL method are that the subgroup symbiotic evolution method uses the subgroup-based population to evaluate the fuzzy rules locally and the adopted immune system algorithm can accelerate the search and increase global search capacity. Finally, the simulation results have shown that the proposed CNFN-ISEL can outperform other methods. © 2011 IEEE.
Wei, C.S., Ko, L.W., Chuang, S.W., Jung, T.P. & Lin, C.T. 2011, 'Genetic feature selection in EEG-based motion sickness estimation', Proceedings of the International Joint Conference on Neural Networks, pp. 365-369.
View/Download from: Publisher's site
Motion sickness is a common symptom that occurs when the brain receives conflicting information about the sensation of movement. Many motion sickness biomarkers have been identified, and electroencephalogram (EEG)-based motion sickness level estimation was found feasible in our previous study. This study employs genetic feature selection to find a subset of EEG features that can further improve estimation performance over the correlation-based method reported in the previous studies. The features selected by genetic feature selection were very different from those obtained by correlation analysis. Results of this study demonstrate that genetic feature selection is a very effective method to optimize the estimation of motion-sickness level. This demonstration could lead to a practical system for noninvasive monitoring of the motion sickness of individuals in real-world environments. © 2011 IEEE.
Huang, C.S., Ko, L.W., Lu, S.W., Chen, S.A. & Lin, C.T. 2011, 'A vectorcardiogram-based classification system for the detection of Myocardial infarction', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 973-976.
View/Download from: Publisher's site
Myocardial infarction (MI), generally known as a heart attack, is one of the top leading causes of mortality in the world. In clinical diagnosis, cardiologists generally utilize 12-lead ECG system to classify patients into MI symptoms: 1. ST segment elevation, 2. ST segment depression or T wave inversion. However unstable ischemic syndromes have rapidly changing supply versus demand characteristics that is one of the several limitations of 12-lead ECG system for MI detection. In addition, the ECG sensor placements of 12-lead system is not easily donned and doffed for tele-healthcare monitoring at home. Vectorcardiogram (VCG) system in clinic is another type of diagnosis plot which represents the magnitude and direction of the electrical potential in the form of a vector loop during cardiac electric activity. The VCG system can easily acquire three ECG waves from X, Y, Z directions to composite vector signal in space and the VCG signals can be transferred to 12-lead ECG signal through Dower transformation and vice versa. Hence, this study attempts to develop a VCG-based classification system for the detection of Myocardial infarction. In the experiment results, the proposed system can select the proper ECG features based on cardiologist's knowledge and proposed principal moments of QRS complex. The classification performance of MI detection can be reached to 99.89% of sensitivity, 92.51% of specificity, 95.35% of positive predictive value, and 96.96% overall accuracy with maximum-likelihood classifier (MLC). © 2011 IEEE.
Lin, C.T., Lin, C.L., Chiu, T.W., Duann, J.R. & Jung, T.P. 2011, 'Effect of respiratory modulation on relationship between heart rate variability and motion sickness', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 1921-1924.
View/Download from: Publisher's site
This study investigates the interplay among heart rate variability (HRV), respiration, and the severity of motion sickness (MS) in a realistic passive driving task. Although HRV is a commonly used metrically in physiological research or even believed to be a direct measure of sympathovagal activities, the results of MS-effected HRV remain mixed across studies. The goal of this study is to find the source of these contradicting results of HRV associated with MS. Experimental results of this study showed that the group trend of the low-frequency (LF) component and the LF/HF ratio increased and high-frequency (HF) component decreased significantly as self-reported MS level increased (p0.001), consistent with a perception-driven autonomic response of the cardiovascular system. However, in one of the subjects, the relationship was reversed when individuals intentionally adjust themselves (deep breathing) to relieve the discomfort of MS during the experiments. It appears that the correlations between HRV and MS level were higher when individuals made fewer adjustments (the number of deep breathing) during the passive driving experiments. © 2011 IEEE.
Chou, Y.H., Chuang, C.C., Zao, J.K., Ko, L.W. & Lin, C.T. 2011, 'An fMRI study of abrupt-awake episodes during behavioral microsleeps', Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, pp. 5060-5063.
View/Download from: Publisher's site
This paper reports the brain activation patterns of five subjects who were abruptly awakened from microsleeps in a simulated automotive driving experiment. By comparing the BOLD signals between behavioral microsleep (BM), abrupt awakening (AA) and post-abrupt awakening (post-AA) stages, we observed that visual area, frontal cortex, limbic lobe manifested more intense activation during the AA stage while frontal cortex, temporal cortex, primary motor area and insula were more activated during the post-AA stage. These results suggested that the subjects were likely in mental states differ from those associated with decision making processes as they went through and emerged from the abrupt awakening episodes. © 2011 IEEE.
Lin, C.T., Li, D.L. & Chang, J.Y. 2011, 'Self-adjusting feature maps network', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 356-364.
View/Download from: Publisher's site
In this paper, we propose a novel artificial neural network, called self-adjusting feature map (SAM), and its unsupervised learning algorithm with self-adjusting mechanism. After the training of SAM network, we will obtain a map composed of a set of representative connected neurons. The trained network structure of representative connected neurons not only displays the spatial relation of the input data distribution but also quantizes the data well. SAM can automatically isolate a set of connected neurons, in which the number of the set may indicate the number of clusters to be used. The idea of self-adjusting mechanism is based on combining of mathematical statistics and neurological advance and retreat of waste. For each representative neuron, there are three periods, growth, adaptation and decline, in its training process. The network of representative neurons will first create the necessary neurons according to the local density of the input data in the growth period. Then it will adjust neighborhood neuron pair's connected/disconnected topology constantly according to the statistics of input feature data in the adaptation period. Lastly the unnecessary neurons of the network will be merged or deleted in the decline period. In this study, we exploit SAM to handle some peculiar cases that cannot be well dealt with by classical unsupervised learning networks such as self-organizing feature map (SOM) network. Furthermore, we also take several real world cases to exhibit the remarkable characteristics of SAM. © 2011 Springer-Verlag.
Ko, L.W., Wei, C.S., Chen, S.A. & Lin, C.T. 2011, 'EEG-based motion sickness estimation using principal component regression', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 717-724.
View/Download from: Publisher's site
Driver's cognitive state monitoring system has been implicated as a causal factor for the safety driving issue, especially when the driver fell asleep or distracted in driving. However, the limitation in developing this system is lack of a major indicator which can be applied to a realistic application. In our past studies, we investigated the physiological changes in the transition of driver's cognitive state by using EEG power spectrum analysis and found that the features in the occipital area were highly correlated with the driver's driving performance. In this study, we construct an EEG-based self-constructed neural fuzzy system to estimate the driver's cognitive state by using the EEG features from the occipital area. Experimental results show that the proposed system had the better performance than other neural networks. Moreover, the proposed system can not only be limited to apply to individual subjects but also sufficiently works in between subjects. © 2011 Springer-Verlag.
Li, C.H., Chu, H.S., Kuo, B.C. & Lin, C.T. 2011, 'Hyperspectral image classification using spectral and spatial information based linear discriminant analysis', International Geoscience and Remote Sensing Symposium (IGARSS), pp. 1716-1719.
View/Download from: Publisher's site
Feature extraction plays an essential role in Hyperspectral image classification. Linear discriminant analysis (LDA) is a commonly used feature extraction (FE) method to resolve the Hughes phenomenon for classification. The Hughes phenomenon (also called the curse of dimensionality) is often encountered in classification when the dimensionality of the space grows and the size of the training set is fixed, especially in the small sampling size problem. Recent studies show that the spatial information can greatly improve the classification performance. Hence, for hyperspectral image classification, it is not only necessary to use the available spectral information but also to exploit the spatial information. In this paper, spatial information is acquired by the concept of the Markov random field (MRF), and this spatial information is used to form the membership values of every pixel in the hyperspectral image. The experimental results on two hyperspectral images, the Washington DC Mall and the Indian Pine Site, show that the proposed method can yield a better classification performance than LDA in the small sampling size problem. © 2011 IEEE.
Lin, C.T., Wang, Y.K. & Chen, S.A. 2011, 'An EEG-based brain-computer interface for dual task driving detection', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 701-708.
View/Download from: Publisher's site
A novel detective model for driver distraction was proposed in this study. Driver distraction is a significant cause of traffic accidents during these years. To study human cognition under a specific driving task, one virtual reality (VR)-based simulation was built. Unexpected car deviations and mathematics questions with stimulus onset asynchrony (SOA) were designed. Electroencephalography (EEG) is a good index for the distraction level to monitor the effects of the dual tasks. Power changing in Frontal and Motor cortex were extracted for the detective model by independent component analysis (ICA). All distracting and non-distracting EEG epochs could be revealed the existence by self-organizing map (SOM). The results presented that this system approached about 90% accuracy to recognize the EEG epochs of non-distracting driving, and might be practicable for daily life. © 2011 Springer-Verlag.
Lin, C.T., Chen, S.A., Ko, L.W. & Wang, Y.K. 2011, 'EEG-based brain dynamics of driving distraction', Proceedings of the International Joint Conference on Neural Networks, pp. 1497-1500.
View/Download from: Publisher's site
Distraction during driving has been recognized as a significant cause of traffic accidents. The aim of this study is to investigate Electroencephalography (EEG) -based brain dynamics in response to driving distraction. To study human cognition under specific driving tasks in a simulated driving experiment, this study utilized two simulated events including unexpected car deviations and mathematics questions. The raw data were first separated into independent brain sources by Independent Component Analysis. Then, the EEG power spectra were used to evaluate the time-frequency brain dynamics. Results showed that increases of theta band and beta band power were observed in the frontal cortex. Further analysis demonstrated that reaction time and multiple cortical EEG power had high correlation. Thus, this study suggested that the features extracted by EEG signal processing, which were the theta power increases in frontal area, could be used as the distracted indexes for early detection of driver inattention in real driving. © 2011 IEEE.
Tseng, H.C., Shyu, J.J., Chang, J.Y. & Lin, C.T. 2011, 'Exploiting automatic image segmentation to human detection and depth estimation', IEEE SSCI 2011 - Symposium Series on Computational Intelligence - CIMSIVP 2011: 2011 IEEE Symposium on Computational Intelligence for Multimedia, Signal and Vision Processing, pp. 19-25.
View/Download from: Publisher's site
In this paper, we combine image segmentation techniques and face detection methods to extract the human from scenes. Firstly, skin regions are detected and an ellipse fitting method is employed to detect the face region and consequently locate the human position. Then we propose an improved automatic seeded region growing algorithm to segment the image. The initial seeds are generated automatically, and the remaining pixels are classified to the nearest region. After the region growing procedure, two neighboring regions with high similarity are merged. The human body is determined by confining semantic human body region in segmented regions, and those belonging to the human face and human body are merged afterward. Lastly, we will detect the human vertical y-coordinate values in the image, and the depths can then be estimated according to the depth look-up tables of the camera. © 2011 IEEE.
Wei, C.S., Ko, L.W., Chuang, S.W., Jung, T.P. & Lin, C.T. 2011, 'EEG-based evaluation system for motion sickness estimation', 2011 5th International IEEE/EMBS Conference on Neural Engineering, NER 2011, pp. 100-103.
View/Download from: Publisher's site
Motion sickness is a common symptom which occurs when the brain receives conflicting sensory information. Although many motion sickness-related biomarkers have been identified, estimating humans' motion sickness level (MSL) remains a challenge in operational environments. Traditionally, questionnaire and physical check are the common ways to passively evaluate subject's sickness level. This study proposes a physiology-based estimation system that can automatically assess subject's motion-sickness level in operational environments. Our previous study showed that increases in self-reported MSL in a Virtual-reality based driving experiment on a motion platform were accompanied by elevated alpha (8-12Hz) power most prominently in the occipital midline electroencephalogram (EEG). This study explores the feasibility of an automatic MSL estimation based on spontaneous EEG spectrum. To this end, this study employed three different estimators: 1) Linear regression (LR), 2) Radial basis function neural network (RBFNN), and 3) Support vector regression (SVR). The results of this study showed that SVR outperformed LR and RBFNN in estimating MSL from EEG spectrum. The averaged accuracy of MSL estimation by SVR was 86.926.09% across 6 subjects. This demonstration could lead to a practical system for noninvasive monitoring of the motion sickness in real-world environments. © 2011 IEEE.
Ko, L.W., Wei, C.S., Jung, T.P. & Lin, C.T. 2011, 'Estimating the level of motion sickness based on EEG spectra', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 169-176.
View/Download from: Publisher's site
Motion sickness (MS) is a normal response to real, perceived, or even anticipated movement. People tend to get motion sickness on a moving boat, train, airplane, car, or amusement park rides. Although many motion sickness-related biomarkers have been identified, but how to estimate human's motion sickness level (MSL) is a big challenge in the operational environment. Traditionally, questionnaire and physical check are the common ways to passively evaluate subject's sickness level. Our past studies had investigated the EEG activities correlated with motion sickness in a virtual-reality based driving simulator. The driving simulator comprised an actual automobile mounted on a Stewart motion platform with six degrees of freedom, providing both visual and vestibular stimulations to induce motion-sickness in a manner that is close to that in daily life. EEG data were acquired at a sampling rate of 500 Hz using a 32-channel EEG system. The acquired EEG signals were analyzed using independent component analysis (ICA) and time-frequency analysis to assess EEG correlates of motion sickness. Subject's degree of motion-sickness was simultaneously and continuously reported using an onsite joystick, providing non-stop psychophysical references to the recorded EEG changes. We found that the parietal, motor, occipital brain regions exhibited significant EEG power changes in response to vestibular and visual stimuli. Based on these findings and experimental results, this study aims to develop an EEG-based system to estimate subject's motion sickness level upon the EEG power spectra from motion-sickness related brain areas. The MS evaluation system can be applied to early detection of the subject's motion sickness and prevent its uncomfortable syndromes in our daily life. Furthermore, the experiment results could also lead to a practical human-machine interface for noninvasive monitoring of motion sickness of drivers or passengers in real-world environments. © 2011 Springer-Verla...
Liao, L.D., Chen, Y.Y., Lin, C.T., Chang, J.Y. & Li, M.L. 2011, 'Functional transcranial photoacoustic micro-imaging of mouse cerebrovascular cross-section and hemoglobin oxygenation changes during forepaw electrical stimulation', Progress in Biomedical Optics and Imaging - Proceedings of SPIE.
View/Download from: Publisher's site
In this study, we report on using a 50-MHz functional photoacoustic microscopy (PAM) to transcranially image the cross-section and hemoglobin oxygenation (SO2) changes of single mouse cortical vessels in response to left forepaw electrical stimulation. Three difference levels of the cortical vessels (i.e., with different-sized diameters of 350, 100 and 55 m) on activated regions were marked to measure their functional cross-section and SO2 changes as a function of time. Electrical stimulation of the mouse left forelimb was applied to evoke functional changes in vascular dynamics of the mouse somatosensory cortex. The applied current pulses were with a pulse frequency of 3 Hz, pulse duration of 0.2 ms, and pulse amplitude of 2 mA. The cerebrovascular cross-section changes, which indicate changes in cerebral blood volume (CBV), were probed by images acquired at 570 nm, a hemoglobin isosbestic point, while SO2 changes were monitored by the derivatives of 560-nm images normalized to 570-nm ones. The results show that vessel diameter and SO2 were significantly dilated and increased when compared with those of the controlled ones. In summary, the PAM shows its promise as a new imaging modality for transcranially functional quantification of single vessel diameter (i.e., CBV) and SO2 changes without any contrast agents applied during stimulation. © 2011 SPIE.
Lin, C.T., Wang, W.R., Wang, I.J., Liao, L.D., Chen, S.F., Tseng, K. & Ko, L.W. 2011, 'A new design of the multi-channels mobile and wireless EEG system', Communications in Computer and Information Science, pp. 293-298.
View/Download from: Publisher's site
Most researchers acquired EEG by using standard measurement system like NeuroScan system, which includes AgCl electrode cap, SynAmps Amplifier and Scan software to provide good reliability for the acquisition of EEG data. However, it is still not convenient for Brain Computer Interface (BCI) application in daily life because of needing conduction gels to contact skins and being wired, expensive and heavy. Moreover, the conduction gel will trend to be drying, so it does not suitable for long-term monitoring. In this study, we developed a mobile and wireless EEG system. The system consists of frond-end 16-channel dry electrode cap, a miniature low-power wireless portable circuitry, and a back-end program receiving events and digital EEG data simultaneously. We demonstrate the recorded EEG data have high correlations between from our system and from NeuroScan system. © 2011 Springer-Verlag.
Chang, J.Y., Lin, Y.Y., Han, M.F. & Lin, C.T. 2011, 'A functional-link based interval type-2 compensatory fuzzy neural network for nonlinear system modeling', IEEE International Conference on Fuzzy Systems, pp. 939-943.
View/Download from: Publisher's site
In this paper, the Functional-Link based Interval Type-2 Compensatory Fuzzy Neural Network (FLIT2CFNN) is a six-layer structure, which combines compensatory fuzzy reasoning method, and the consequent part is combined the proposed functional-link neural network with interval weights. The compensatory fuzzy reasoning method uses adaptive fuzzy operations of neuro-fuzzy systems that can make the fuzzy logic system more adaptive and effective. Initially, there is no rule in the FLIT2CFNN. A FLIT2CFNN is constructed using concurrent structure and parameter learning. The advantages of this learning algorithm are that it converges quickly and the obtained fuzzy rules are more precise. All of the antecedent part parameters and compensatory degree values are learned by gradient descent algorithm. Several simulation results show that the FLIT2CFNN achieves better performance than other feedforword type-1 and type-2 FNNs. © 2011 IEEE.
Chu, H.S., Kuo, B.C., Li, C.H. & Lin, C.T. 2011, 'A semisupervised feature extraction method based on fuzzy-type linear discriminant analysis', IEEE International Conference on Fuzzy Systems, pp. 1927-1932.
View/Download from: Publisher's site
Linear discriminant analysis (LDA) is a commonly used feature extraction (FE) method to resolve the Hughes phenomenon for classification. The Hughes phenomenon (also called the curse of dimensionality) is often encountered in classification when the dimensionality of the space grows and the size of the training set is fixed, especially in the small sampling size problem. Recent studies show that the spatial information can greatly improve the classification performance. Hence, for hyperspectral image classification, it is not only necessary to use the available spectral information but also to exploit the spatial information. In this paper, a semisupervised feature extraction method which is based on the scatter matrices of the fuzzy-type LDA and uses the semi-information is proposed. The experimental results on two hyperspectral images, the Washington DC Mall and the Indian Pine Site, show that the proposed method can yield a better classification performance than LDA in the small sampling size problem. © 2011 IEEE.
Lin, C.T., Han, M.F., Lin, Y.Y., Liao, S.H. & Chang, J.Y. 2011, 'Neuro-fuzzy system design using differential evolution with local information', IEEE International Conference on Fuzzy Systems, pp. 1003-1006.
View/Download from: Publisher's site
This paper proposes a differential evolution with local information for TSK-type neuro-fuzzy system optimization. The differential evolution with local information consider neighborhood between each individual to keep the diversity of population. An adaptive parameter tuning based on 1/5th rule is used to trade off between local search and global search. For structure learning algorithm, the on-line clustering algorithm is used for rule generation. The structure learning algorithm generates a new rule which compares the firing strength. Initially, there is no rule in neuro-fuzzy system model. The rules are automatically generated by fuzzy measure. For parameter learning, the parameters are optimized by differential evolution algorithm. Finally, the proposed neuro-fuzzy system with novel differential evolution model is applied in chaotic sequence prediction problem. Results of this paper demonstrate the effectiveness of the proposed model. © 2011 IEEE.
Li, C.H., Lin, C.T., Kuo, B.C. & Chu, H.S. 2010, 'An automatic method for selecting the parameter of the RBF kernel function to support vector machines', International Geoscience and Remote Sensing Symposium (IGARSS), pp. 836-839.
View/Download from: Publisher's site
Support vector machine (SVM) is one of the most powerful techniques for supervised classification. However, the performances of SVMs are based on choosing the proper kernel functions or proper parameters of a kernel function. It is extremely time consuming by applying the k-fold cross-validation (CV) to choose the almost best parameter. Nevertheless, the searching range and fineness of the grid method should be determined in advance. In this paper, an automatic method for selecting the parameter of the RBF kernel function is proposed. In the experimental results, it costs very little time than k-fold cross-validation for selecting the parameter by our proposed method. Moreover, the corresponding SVMs can obtain more accurate or at least equal performance than SVMs by applying k-fold cross-validation to determine the parameter. © 2010 IEEE.
Jung, T.P., Huang, K.C., Chuang, C.H., Chen, J.A., Ko, L.W., Chiu, T.W. & Lin, C.T. 2010, 'Arousing feedback rectifies lapse in performance and corresponding EEG power spectrum', 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC'10, pp. 1792-1795.
View/Download from: Publisher's site
This study explores electroencephalographic (EEG) dynamics and behavioral changes in response to arousing auditory signals presented to individuals experiencing momentary cognitive lapses. Arousing auditory feedback was delivered to the subjects in half of the non-responded lane-deviation events during a sustained-attention driving task, which immediately agitated subject's responses to the events. The improved behavioral performance was accompanied by concurrent power suppression in the theta-and alpha-bands in the lateral occipital cortices. This study further explores the feasibility of estimating the efficacy of arousing feedback presented to the drowsy subjects by monitoring the changes in EEG power spectra. © 2010 IEEE.
Wang, I.J., Liao, L.D., Wang, Y.T., Chen, C.Y., Lin, B.S., Lu, S.W. & Lin, C.T. 2010, 'A wearable mobile electrocardiogram measurement device with novel dry polymer-based electrodes', IEEE Region 10 Annual International Conference, Proceedings/TENCON, pp. 379-384.
View/Download from: Publisher's site
A Wearable Mobile Electrocardiogram Monitoring System (WMEMS), which mainly consists of a wearable Electrocardiogram (ECG) acquisition device, a mobile phone with global positioning system, and a healthcare server, was developed in this study. Most of telemedicine systems for long-term ECG monitoring focus on the application of communication techniques. However, how to monitor long-term ECG state more comfortably in daily life is also an important issue. In this study, a novel dry foam electrode was designed and applied for the wearable ECG acquisition device in our WMEMS. These novel dry foam electrodes without conduction gels can provide good conductivity to acquire ECG signal effectively, and can adapt to irregular skin surface to maintain low skin-electrode impedance and reduce motion artifacts under movement. Therefore, the wearable ECG acquisition device is suitable for long-term ECG monitoring in daily life. Moreover, by combining with wireless communication technique, our WMEMS can monitor patient's heart rate continuously anywhere in the globe if they are under the coverage of GSM cellular network. Experiment results showed that our WMEMS really provides a good system prototype for ECG telemedicine applications. © 2010 IEEE.
Han, M.F., Lin, C.T. & Chang, J.Y. 2010, 'A compensatory neurofuzzy system with online constructing and parameter learning', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 552-556.
View/Download from: Publisher's site
A compensatory neurofuzzy system (CNFS) with on-line learning ability is proposed in this paper. The proposed CNFS model uses a compensatory layer to raise the diversity of fuzzy rules by compensatory weights. The compensatory layer can automatically compare with each fuzzy rule and select higher resources for more important fuzzy rule. An online learning algorithm, which consists of structure learning and parameter learning, is also presented. The structure learning depends on the fuzzy measure to determine the number of fuzzy rules. The parameter learning, based on the gradient descent method, can adjust the shape of the membership function and the weights of the compensatory layer. To demonstrate the capability of the proposed CNFS, it is applied to the Iris, and Wisconsin breast cancer classification datasets from the VCI Repository. Experimental results show that the proposed CNFS for pattern classification can achieve good classification performance. ©2010 IEEE.
Liao, S.H., Lin, C.T. & Chang, J.Y. 2010, 'Preliminary study on additive radial basis function networks', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 3113-3117.
View/Download from: Publisher's site
In this paper, a new class of learning models, namely the additive radial basis function networks (ARBFNs) for general nonlinear regression problems are proposed. This class of learning machines combines the radial basis function networks (RBFNs) commonly used in general machine learning problems and the additive models (AMs) frequently encountered in semiparametric regression problems. In statistical regression theory, AM is a good compromise between the linear parametric model and the nonparametric model. Simulation results show that for the given learning problem, ARBFNs usually need fewer hidden nodes than those of RBFNs for the same level of accuracy. ©2010 IEEE.
Wang, Y.K., Pal, N.R., Lin, C.T. & Chen, S.A. 2010, 'Analyzing effect of distraction caused by dual-tasks on sharing of brain resources using SOM', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: Publisher's site
Drivers' distraction is widely recognized as a leading cause of car accidents. To investigate the distracting effect of dual-tasks involving driving and answering mathematical equations in the stimulus onset asynchrony (SOA) conditions, we design five different cases: two cases involving single-tasks and three cases involving dual-tasks. We have found that there is no statistically significant change in the behavioral data among the three dual-tasks. This raises an important question - is there any detectable effect of the dual tasks on the brain waves? To answer this, we use the Self-Organizing Map (SOM) to recognize the changes, if any, in the Electroencephalography (EEG) dynamics associated with such dual-tasks. Our SOM analysis based on independent components corresponding to EEG signals extracted from Frontal and Motor areas revealed that single- and dual-tasks have distinguishable signatures in the EEG signals. Specifically, each of the two single-task conditions is clustered in a distinct spatial area of the map. Two of the dual-tasks also exhibit distinct spatial clusters, while the third case although shows differences from the other two, the neurons corresponding to this case are sub-clustered reflecting the fact that different subjects may give different priorities to the tasks when confronted with two tasks simultaneously. SOM-based exploratory analysis reveals the existence of distinct EEG signatures among the distracting and non-distracting tasks, although there is no any noticeable difference in the behavioral data among these cases. © 2010 IEEE.
Han, M.F., Liao, L.D., Liu, Y.H., Wang, W.R., Lin, B.S. & Lin, C.T. 2010, 'Performance optimized of the novel dry EEG electrodes by using the Non-Dominated Sorting Genetic Algorithms (NSGA-II)', IEEE Region 10 Annual International Conference, Proceedings/TENCON, pp. 1710-1715.
View/Download from: Publisher's site
In this study, a optimization process was performed for the developed dry electroencephalography (EEG) electrodes by using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to minima the skin-electrode impedance. The developed dry EEG electrodes can measure the EEG signals without any gels applied and no skin preparation. However, how to find a proper skin-electrode contact area is an important issue. The contact area is directly related to the electrodes impedance and fabrication cost. Therefore, the NSGA-II is used to searching the suitable contact area and other design parameters. NSGA-II is a wieldy used optimization method, especially for the multi-objectives issues like this case. Finally, we compare the results of the simulation and experiments for ensuring the optimal process. The experiment results show that using the optimal values provided from NSGA-II can achieve the minima skin-electrode impedance. It confirms the dry electrode can be effectively used for the cognitive or other applications in the future. ©2010 IEEE.
Lin, Y.Y., Chang, J.Y. & Lin, C.T. 2010, 'An internal/interconnection recurrent type-2 fuzzy neural network (IRT2FNN) for dynamic system identification', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 733-737.
View/Download from: Publisher's site
This paper proposes an Internal/Interconnection Recurrent Type-2 Fuzzy Neural Network (IRT2FNN) for dynamic system identification. The antecedent part of IRT2FNN forms a self and interconnection feedback loop by feeding the past and current firing strength of each rule . The TSK-type consequent part is a linear model of exogenous inputs with interval weights. The initial rule base in the IRT2FNN is empty, and an on-line constructing method is proposed to generate fuzzy rules which flexibly partition the input space. The recurrent structure in the IRT2FNN enable to handle dynamic system identification problems with a priori knowledge of system input and output delay numbers. Simulations on dynamic system identification verify the performance of IRT2FNN with clean and noisy outputs as well. ©2010 IEEE.
Hung, S.H., Chao, C.F., Yan, Y.C., Lin, B.S. & Lin, C.T. 2010, 'Independent component analysis hard-IP integration System on Programmable Chip (SOPC) platform', IEEE Region 10 Annual International Conference, Proceedings/TENCON, pp. 1705-1709.
View/Download from: Publisher's site
This paper presented Independent Component Analysis (ICA) Hard-IP integration in System on Programmable Chip (SOPC) platform. The ICA component can discover the main component for original signal in multiple fetching signal sources, and it has been used in biomedical signal processing such as electroencephalogram (EEG) analysis. The proposed system consists of a programmable CPU, ICA processing units, system bus, communication, and display interface. The experimental results showed that the proposed design implemented on Altera DE2 FPGA development board, can achieve real-time signal separation and display at 100 MHz. The whole design consists of 29,640 logic elements. ©2010 IEEE.
Khoo, I.H., Reddy, H.C., Van, L.D. & Lin, C.T. 2010, 'Generalized formulation of 2-D filter structures without global broadcast for VLSI implementation', Midwest Symposium on Circuits and Systems, pp. 426-429.
View/Download from: Publisher's site
A generalized formulation is developed that allows the derivation of various new 2-D VLSI filter structures, without global broadcast, using different filter sub-blocks and their interconnections (frameworks). With this formulation, lattice-type and direct-form structures realizing general 2-D IIR and FIR transfer functions, IIR transfer functions with separable denominators, and transfer functions with quadrantal magnitude symmetry are easily obtained. The separable denominator and quadrantal symmetry structures have the advantage of reduced number of multipliers. © 2010 IEEE.
Lin, F.C., Ko, L.W., Chen, S.A., Chen, C.F. & Lin, C.T. 2010, 'EEG-based cognitive state monitoring and predition by using the self-constructing neural fuzzy system', ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, pp. 2287-2290.
View/Download from: Publisher's site
Driver's cognitive state monitoring has been implicated as a causal factor for the safety driving issue, especially when the driver fell asleep or distracted in driving. In our past studies, we found that the EEG power spectrum changes were highly correlated with the driver's driving behavior performance. In this study, we attempt to construct an EEG-based self-constructing neural fuzzy system to monitor and predict the driver's cognitive state. The difficulties in developing such a system are lack of significant index for detecting drowsiness and the interference of the complicated noise in a realistic and dynamic driving environment. Our experimental results including correlation and prediction show that the performances of our proposed system are significantly higher than using the traditional neural networks. Besides, the proposed EEG-based self-constructing neural fuzzy system can be generalized and applied in the subjects' independent sessions. This unique advantage can be widely used in the real-life applications. ©2010 IEEE.
Chen, P.H. & Lin, C.T. 2010, 'Sequential clustering by triangle-cascaded robot deployment', Proceedings of the IEEE International Conference on Industrial Technology, pp. 597-601.
View/Download from: Publisher's site
A virtual robot deploys its joints and linkages step by step in a 2-D region with presented obstacles. Each step of deployment constructs a piece of virtual robot trajectory based on only a few obstacles in front. The virtual robot trajectory serves as an envelope for obstacle clusters. Sequential clustering is thus called to solve this issue. The innovation of triangle cascading, composed of joint discrimination and apex least-square deployment, reflects the idea of sequential clustering. Simulation covers triangle cascading, gap comparison and common rim or common apex for further deployment, link and reduction of joint trajectory. An alternative test pattern using random-distributed obstacles validates algorithms developed in this paper. A hybrid clustering combining fuzzy c-means and hierarchical clustering shows a qualified approach for the validation eventually. ©2010 IEEE.
Chuang, C.H., Lai, P.C., Ko, L.W., Kuo, B.C. & Lin, C.T. 2010, 'Driver's cognitive state classification toward brain computer interface via using a generalized and supervised technology', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: Publisher's site
Growing numbers of traffic accidents had become a serious social safety problem in recent years. The main factor of the high fatalities was the obvious decline of the driver's cognitive state in their perception, recognition and vehicle control abilities while being sleepy. The key to avoid the terrible consequents is to build a detecting system for ongoing assessment of driver's cognitive state. A quickly growing research, brain-computer interface (BCI), offers a solution offering great assistance to those who require alternative communicatory and control mechanisms. In this study, we propose an alertness/drowsiness classification system based on investigating electroencephalographic (EEG) brain dynamics in lane-keeping driving experiments in a virtual reality (VR) driving environment with a motion platform. The core of the classification system is composed of dimension reduction technique and classifier learning algorithm. In order to find the suitable method for better describing the data structure, we explore the performances using different feature extraction and feature selection methods with different classifiers. Experiment results show that the accuracy is over 80% in most combinations and even near 90% under Principal Component Analysis (PCA) and Nonparametric Weighted Feature Extraction (NWFE) going with Gaussian Maximum Likelihood classifier (ML) and k-Nearest-Neighbor classifier (kNN), respectively. In addition, this developed classification system can also solve the individual brain dynamic differences caused from different subjects and overcome the subject dependent limitation. The optimized solution with better accuracy performance out of all combinations can be considered to implement in the kernel brain-computer interface. © 2010 IEEE.
Li, C.H., Lin, C.T., Kuo, B.C. & Ho, H.H. 2010, 'An automatic method for selecting the parameter of the normalized kernel function to support vector machines', Proceedings - International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2010, pp. 226-232.
View/Download from: Publisher's site
Support vector machine (SVM) is one of the most powerful techniques for supervised classification. However, the performances of SVMs are based on choosing the proper kernel functions or proper parameters of a kernel function. It is extremely time consuming by applying the k-fold cross-validation (CV) to choose the almost best parameter. Nevertheless, the searching range and fineness of the grid method should be determined in advance. In this paper, an automatic method for selecting the parameter of the normalized kernel function is proposed. In the experimental results, it costs very little time than k-fold cross-validation for selecting the parameter by our proposed method. Moreover, the corresponding SVMs can obtain more accurate or at least equal performance than SVMs by applying k-fold cross-validation to determine the parameter. © 2010 IEEE.
Chen, P.H. & Lin, C.T. 2010, '6-Step fuzzy-merged controller for eccentricity of 3-pole vertical magnetic bearing', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2640-2645.
View/Download from: Publisher's site
Centrifugal force caused by eccentricity in a vertical magnetic bearing (VMB) is dealt with an innovative design by featuring 1)3-pole 2)6-step 3)PD fuzzy switching controller 4)hybrid controller. The hybrid controller is composed of a fuzzy switching controller and a traditional ID controller. Fuzzy switching controller regulates the instability of plant when pump shaft deviates by a small perturbation from the nominal gap. The traditional ID controller compensates for the centrifugal force caused by eccentricity. In the simulation, 6-step switching works as a relay race in terms of a competitor. Simulation results approve the approach developed in this paper be feasible to overcome the effect arisen from eccentricity. ©2010 IEEE.
Yu, Y.H., Lai, P.C., Ko, L.W., Chuang, C.H., Kuo, B.C. & Lin, C.T. 2010, 'An EEG-based classification system of passenger's motion sickness level by using feature extraction/selection technologies', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: Publisher's site
Past studies reported that the main electrogastrography (EEG) dynamic changes related to motion sickness (MS) were occurred in occipital, parietal, and somatosensory brain area, especially in the power increasing of the alpha band (8-13 Hz) and theta band (4-7 Hz) which had positive correlation with the subjective MS level. Depend on these main findings correlated with MS, we attempt to develop an EEG based classification system to automatically classify subject's MS level and find the suitable EEG features via common feature extraction, selection and classifiers technologies in this study. If we can find the regulations and then develop an algorithm to predict MS occurring, it would be a great benefit to construct a safe and comfortable environment for all drivers and passengers when they are cruising in the car, bus, ship or airplane. EEG is one of the best methods for monitoring the brain dynamics induced by motion-sickness because of its high temporal resolution and portability. After collecting the EEG signals and subjective MS level in a realistic driving environment, we first do the data pre-processing part including ICA, component clustering analysis and time-frequency analysis. Then we adopt three common feature extractions and two feature selections (FE/FS) technologies to extract or select the correlated features such as principal component analysis (PCA), linear discriminate analysis (LDA), nonparametric weighted feature extraction (NWFE), forward feature selections (FFS) and backward feature selections (BFS) and feed the feature maps into three classifiers (Gaussian Maximum Likelihood Classifier (ML), k-Nearest-Neighbor Classifier (kNN) and Support Vector Machine (SVM)). Experimental results show that classification performance of all our proposed technologies can be reached almost over 95%. It means it is possible to apply the effective technology combination to predict the subject's MS level in the real life applications. The better combination in ...
Liao, L.D., Wang, I.J., Chang, C.J., Lin, B.S., Lin, C.T. & Tseng, K.C. 2010, 'Human cognitive application by using wearable mobile brain computer interface', IEEE Region 10 Annual International Conference, Proceedings/TENCON, pp. 346-351.
View/Download from: Publisher's site
In this study, we demonstrated a Wearable Mobile Electroencephalogram (EEG)-based Brain Computer Interface System (WMEBCIS) with foam-based dry electrodes embedded. According to the previous studies, many kinds of BCI systems have been proposed and target the applications on sensors and information techniques. For daily life EEG monitoring by using the BCI systems, the sensors have employ a key technique for ensuring the data quality. In view of this, a novel dry foam electrode was designed and applied for the wearable EEG acquisition device in our WMEBCIS. The developed dry electrodes can achieve the quick-placement and high signals quality under no skin preparation and without gels applied. Most importantly, due to its softness advantage, the developed electrodes can fit properly on the skin surface to maintain low skin-electrode impedance and reduce motion artifacts under movement. By using the specific mechanism of our wearable EEG acquisition device, it can be worn conveniently and can maintain the electrode-scalp contact area to acquire good EEG signal quality effectively. Therefore, the wearable EEG acquisition device is suitable for long-term EEG monitoring in daily life. In words, our WMEBCIS was also demonstrated for a cognitive application on drowsiness detection. © 2010 IEEE.
Hung, S.H., Chang, C.J., Chao, C.F., Wang, I.J., Lin, C.T. & Lin, B.S. 2010, 'Development of real-time wireless brain computer interface for drowsiness detection', ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, pp. 1380-1383.
View/Download from: Publisher's site
In this study, a real-time wireless embedded EEGbased brain computer interface (BCI) system was developed for drowsiness detection in a realistic driving task. Accidents caused by driver's drowsiness behind the steering wheel have a high fatality rate because of the marked decline in the driver's abilities of perception, recognition, and vehicle control abilities while sleepy. Therefore, real-time drowsiness monitoring is important to avoid traffic accidents. In this study, an embedded EEG-based BCI system which includes a wireless physiological signal acquisition module and an embedded signal processing module was designed, and a real-time drowsiness detection algorithm based on our unsupervised approach was implemented in the embedded signal processing module. EEG signal would be monitored and analyzed by the embedded signal processing module, and the warning tone would be triggered to prevent traffic accidents when the drowsiness condition occurred. ©2010 IEEE.
Lin, C.T., Lin, C.L., Huang, K.C., Chen, S.A. & Tung, J.H. 2010, 'The performance of visuo-motor coordination changes under force feedback assistance system', ISCAS 2010 - 2010 IEEE International Symposium on Circuits and Systems: Nano-Bio Circuit Fabrics and Systems, pp. 1376-1379.
View/Download from: Publisher's site
In this study, a system with force feedback assistance was adopted to improve human's motor learning. Subjects tried to perform a trajectory tracking task of visuo-motor coordination by using a joystick which involved force feedback assistance. Force feedback was proportionally generated (maximal output 9N) by the joystick and warned subjects of the deviation of the tracking task. In the experiments, motor behavior performances of assistance group and non-assistance group were taken into comparison. The results demonstrated that a joystick with force feedback assistance brought significant influence. Motor performance was improved in the motion of flexion and extension while adding the force feedback assistance. This study provided important new light on the force feedback assistance. It could be an advanced function to improve and deepen people's motor learning capabilities. ©2010 IEEE.
Jung, T.P., Huang, K.C., Chuang, C.H., Chen, J.A., Ko, L.W., Chiu, T.W. & Lin, C.T. 2010, 'Arousing feedback rectifies lapse in performance and corresponding EEG power spectrum.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, pp. 1792-1795.
This study explores electroencephalographic (EEG) dynamics and behavioral changes in response to arousing auditory signals presented to individuals experiencing momentary cognitive lapses. Arousing auditory feedback was delivered to the subjects in half of the non-responded lane-deviation events during a sustained-attention driving task, which immediately agitated subject's responses to the events. The improved behavioral performance was accompanied by concurrent power suppression in the theta- and alpha-bands in the lateral occipital cortices. This study further explores the feasibility of estimating the efficacy of arousing feedback presented to the drowsy subjects by monitoring the changes in EEG power spectra.
Duann, J.R., Chen, P.C., Ko, L.W., Huang, R.S., Jung, T.P. & Lin, C.T. 2009, 'Detecting frontal EEG activities with forehead electrodes', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 373-379.
View/Download from: Publisher's site
This study demonstrates the acquisitions of EEG signals from non-hairy forehead sites and tested the feasibility of using the forehead EEG in detecting drowsiness-related brain activities. A custom-made 15-channel forehead EEG-electrode patch and 28 scalp electrodes placed according to the International 10-20 system were used to simultaneously record EEG signals from the forehead and whole-head regions, respectively. A total of five subjects were instructed to perform a night-time long-haul driving task for an hour in a virtual-reality based driving simulator comprising a real car mounted on a 6 degree-of-freedom Steward motion platform and a immersive VR environment with 360 degree projection scenes. Separate independent component analyses were applied to the forehead and whole-head EEG data for each individual subject. For the whole-head independent component (IC) set, the frontal central midline (FCM) IC with an equivalent dipole source located in the anterior cingulate cortex was selected for further analysis. For the forehead IC set, the IC with its theta power changes highly correlated with subject's driving performance was selected. The EEG power changes of the selected forehead ICs were then used to predict driving performance based on a linear regression model. The results of this study showed that it is feasible to accurately estimate quantitatively the changing level of driving performance using the EEG features obtained from the forehead non-hairy channels, and the estimation accuracy was comparable to that using the EEG features of the whole-head recordings. © 2009 Springer.
Liang, S.F., Liu, C.S., Chang, W.L., Tsao, Y.H., Ko, L.W. & Lin, C.T. 2009, 'Assessment of musical training induced neuroplasticity by auditory event related potentials and neural networks', Proceedings of the International Joint Conference on Neural Networks, pp. 1797-1801.
View/Download from: Publisher's site
Music provides a tool to study numerous aspects of neuroscience from motion-skill to emotion since listening to and producing music involves many brain functions. The musician's brain is also regarded as an ideal model to investigate plasticity of the human brain. In this paper, an EEG-based neural network is proposed to assess neuroplasticity induced by musical training. A musical chord perception experiment is designed to acquire and compare the behavioral and neural responses of musicians and non-musicians. The ERPs elicited by the consonant and dissonant chords are combined together as the features of the model. The principle component analysis (PCA) is used to reduce feature dimensions and the dimension-reduced features are input to a feedforward neural network to recognize the brain potentials belong to a musician or a non-musician. The accuracy can reach 97% in average for leave-one-out cross validation of six subjects in this experiment. It demonstrates the feasibility of assessing effects of musical training by ERP signals elicited by musical chord perception. © 2009 IEEE.
Lin, C.T., Yang, F.S., Chiou, T.C., Ko, L.W., Duann, J.R. & Gramann, K. 2009, 'EEG-based spatial navigation estimation in a virtual reality driving environment', Proceedings of the 2009 9th IEEE International Conference on Bioinformatics and BioEngineering, BIBE 2009, pp. 435-438.
View/Download from: Publisher's site
The aim of this study is to investigate the difference of EEG dynamics on navigation performance. A tunnel task was designed to classify subjects into allocentric or egocentric spatial representation users. Despite of the differences of mental spatial representation, behavioral performance in general were compatible between the two strategies subjects in the tunnel task. Task-related EEG dynamics in power changes were analyzed using independent component analysis (ICA), time-frequency and non-parametric statistic test. ERSP image results revealed navigation performance-predictive EEG activities which is is expressed in the parietal component by source reconstruction. For egocentric subjects, comparing to trails with well-estimation of homing angle, the power attenuation at the frequencies from 8 to 30 Hz (around alpha and beta band) was stronger when subjects overestimated homing directions, but the attenuated power was decreased when subjects were underestimated the homing angles. However, we did not found performance related brain activities for allocentric subjects, which may due to the functional dissociation between the use of allo- and egocentric reference frames. © 2009 IEEE.
Ko, L.W., Tsai, I.L., Yang, F.S., Chung, J.F., Lu, S.W., Jung, T.P. & Lin, C.T. 2009, 'Real-time embedded EEG-based brain-computer interface', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 1038-1045.
View/Download from: Publisher's site
Online artifact rejection, feature extraction, and pattern recognition are essential to advance the Brain Computer Interface (BCI) technology so as to be practical for real-world applications. The goals of BCI system should be a small size, rugged, lightweight, and have low power consumption to meet the requirements of wearability, portability, and durability. This study proposes and implements a moving-windowed Independent Component Analysis (ICA) on a battery-powered, miniature, embedded BCI. This study also tests the embedded BCI on simulated and real EEG signals. Experimental results indicated that the efficacy of the online ICA decomposition is comparable with that of the offline version of the same algorithm, suggesting the feasibility of ICA for online analysis of EEG in a BCI. To demonstrate the feasibility of the wearable embedded BCI, this study also implements an online spectral analysis to the resultant component activations to continuously estimate subject's task performance in near real time. © 2009 Springer Berlin Heidelberg.
Liao, L.D., Chao, P.C.P., Chen, J.T., Chen, W.D., Hsu, W.H., Chiu, C.W. & Lin, C.T. 2009, 'A miniaturized electromagnetic generator with planar coils and its energy harvest circuit', IEEE Transactions on Magnetics, pp. 4621-4627.
View/Download from: Publisher's site
This study presents design, analysis and ex;periment of a miniaturized rotary generator in size of 10 10 2 mm3 and its compact energy harvest circuit chip. The designed generator consists of patterned planar copper coils and a multipolar hard magnet ring made of NdFeB. To perform modeling, a harmonic-like magnetic field model along the circumferential path of each magnetic pole is assumed with the assistance from measured peak magnetic flux densities. This is followed by the application of Faraday's law to predict generated electromotive forces (EMFs) in terms of the relative rotational speed between the magnet ring and coils. The genetic algorithm (GA) is next applied to optimize the critical dimensions of the miniaturized generator. The theoretical model of this power microgenerator is evaluated and compared with experimental results, and it is found that the analytical simulation shows a good agreement with the experimental results. The optimized generator offers 4.5 V and 7.23 mW in root mean square (rms) at 10 000 r/min. With microgenerator successfully fabricated, a novel energy harvest circuit employing Dickson charge pump is designed and fabricated via the 0.35-um process offered by National Chip Implementation Center (CIC) of Taiwan. This charge pump circuit owns the merit of almost-zero thresholds of employed metal-oxide-semiconductor (MOS) transistors, enabling the conversion of low-power alternating current (ac) signals by the microgenerator to direct current (dc) ones. © 2009 IEEE.
Chuang, S.W., Huang, R.S., Ko, L.W., Jeng, J.L., Duann, J.R., Jung, T.P. & Lin, C.T. 2009, 'Independent modulators mediate spectra of multiple brain processes In a VR-based driving experiment', Proceedings of SPIE - The International Society for Optical Engineering.
View/Download from: Publisher's site
This study explores the use of Independent Component Analysis (ICA) applied to normalized logarithmic spectral changes in the activities of brain processes separated by spatial filters learned from electroencephalogram (EEG) data using a temporal ICA. EEG data were collected during 1-2 hour virtual-reality based driving experiments, in which subjects were instructed to maintain their cruising position and compensate for randomly induced drifts using the steering wheel. ICA was first applied to 30-channel EEG data to separate the recorded signals into a sum of maximally temporally independent components (ICs) for each of 15 subjects. Logarithmic spectra of IC activities were then submitted to PCA-ICA to find spectrally fixed and temporally independent modulator (IM) processes. The second ICA detected and modeled independent co-modulatory systems that multiplicatively affect the activities of spatially distinct IC processes. Across subjects, we found two consistent temporally independent modulators: theta-beta and alpha modulators that mediate spectral activations of the distinct cortical areas when the participants experience waves of alternating alertness and drowsiness during long hour simulated driving. Furthermore, the time courses of the theta-beta modulator were highly correlated with concurrent changes in subject driving error (a behavioral index of drowsiness). © 2009 SPIE.
Liao, L.D., Chao, P.C.P., Chen, Y.H., Lin, C.T., Ko, L.W., Lin, H.H. & Hsu, W.H. 2009, 'A novel hybrid bioelectrode module for the zero-prep EEG measurements', Proceedings of IEEE Sensors, pp. 939-942.
View/Download from: Publisher's site
This paper discusses a novel hybrid EEG biosensor module that measures through-hair EEG with zero preparation of the scalp. The biosensors can be donned or doffed quickly by the wearer, requiring no skin preparation, produce no skin irritation, and be comfortably worn for an extended period. The proposed EEG sensor is fabricated by the injection molding process with a flexible base and biofriendly materials. The noninvasive nature of BRCNCTU's EEG system [1] (i.e. zero skin preparation, minimal wiring) permits greater subject freedom of motion and considerably improves user compliance for such systems. This effort culminates in a biomonitoring system, measuring EEG, EOG, ECG and EMG signals for real time monitoring of human-computer interactions. ©2009 IEEE.
Lee, C.Y., Lin, C.T. & Hong, C.T. 2009, 'Spatio-temporal analysis in smoke detection', ICSIPA09 - 2009 IEEE International Conference on Signal and Image Processing Applications, Conference Proceedings, pp. 80-83.
View/Download from: Publisher's site
Smoke detection in video surveillance images has been studied for years. However, given an image in open or large spaces with typical smoke and the disturbance of commonly moving objects such as pedestrians or vehicles, robust and efficient smoke detection is still a challenging problem. In this paper, we present a novel and reliable framework for automatic smoke detection. It exploits three features: edge blurring, the gradual change of energy and the gradual change of chromatic configuration. In order to gain proper generalization ability with respect to sparse training samples, the three features are combined using a support vector machine based classifier. This system has been run more than 6 hours in various conditions to verify the reliability of fire safety in the real world.
Lin, C.T., Ko, L.W., Chang, C.J., Wang, Y.T., Chung, C.H., Yang, F.S., Duann, J.R., Jung, T.P. & Chiou, J.C. 2009, 'Wearable and wireless brain-computer interface and its applications', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 741-748.
View/Download from: Publisher's site
This study extends our previous work on mobile & wireless EEG acquisition to a truly wearable and wireless human-machine interface, NCTU Brain-Computer-Interface-headband (BCI-headband), featuring: (1) dry Micro-Electro-Mechanical System (MEMS) EEG electrodes with 400 ganged contacts for acquiring signals from non-hairy sites without use of gel or skin preparation; (2) a miniature data acquisition circuitry; (3) wireless telemetry; and (4) online signal processing on a commercially available cell phone or a lightweight, wearable digital signal processing module. The applicability of the NCTU BCI-headband to EEG monitoring in real-world environments was demonstrated in a sample study: cognitive-state monitoring and management of participants performing normal tasks. © 2009 Springer.
Chen, Y.C., Duann, J.R., Lin, C.L., Chuang, S.W., Jung, T.P. & Lin, C.T. 2009, 'Motion-sickness related brain areas and EEG power activates', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 348-354.
View/Download from: Publisher's site
This study investigates electroencephalographic (EEG) correlates of motion sickness in a virtual-reality based driving simulator. The driving simulator comprised an actual automobile mounted on a Stewart motion platform with six degrees of freedom, providing both visual and vestibular stimulations to induce motion-sickness in a manner that is close to that in daily life. EEG data were acquired at a sampling rate of 500 Hz using a 32-channel EEG system. The acquired EEG signals were analyzed using independent component analysis (ICA) and time-frequency analysis to assess EEG correlates of motion sickness. Subject's degree of motion-sickness was simultaneously and continuously reported using an onsite joystick, providing non-stop psychophysical references to the recorded EEG changes. Five Motion-sickness related brain processes with equivalent dipoles located in the left motor, the parietal, the right motor, the occipital and the occipital midline areas were consistently identified across all subjects. These components exhibited distinct spectral suppressions or augmentation in motion sickness. The results of this study could lead to a practical human-machine interface for noninvasive monitoring of motion sickness of drivers or passengers in real-world environments. © 2009 Springer.
Khoo, I.H., Reddy, H.C., Van, L.D. & Lin, C.T. 2009, '2-D digital filter architectures without global broadcast and some symmetry applications', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 952-955.
View/Download from: Publisher's site
Four new 2-D filter VLSI architectures without global broadcast are presented. The first is a transposed systolic structure which requires fewer delay elements compared to the original systolic structure in [1]. By combining the sub-blocks of the original with the new transposed structure, two additional systolic structures are obtained to realize transfer functions with separable denominators, which require fewer multipliers. These separable denominator structures have important symmetry applications. A structure which possesses diagonal symmetry is then shown which requires even fewer multipliers. ©2009 IEEE.
Kuo, B.C., Chuang, C.H., Li, C.H. & Lin, C.T. 2009, 'Subspace selection based multiple classifier systems for hyperspectral image classification', WHISPERS '09 - 1st Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing.
View/Download from: Publisher's site
In a typical supervised classification task, the size of training data fundamentally affects the generality of a classifier. Given a finite and fixed size of training data, the classification result may be degraded as the number of features (dimensionality) increase. Many researches have demonstrated that multiple classifier systems (MCS) or socalled ensembles can alleviate small sample size and high dimensionality concern, and obtain more outstanding and robust results than single models. One of the effective approaches for generating an ensemble of diverse base classifiers is the use of different feature subsets such as random subspace method (RSM). The objective of this research is to develop a novel ensemble technique based on cluster algorithms for strengthening RSM. The results of real data experiments show that the proposed method obtains the sound performance especially in the situation of using less number of classifiers. © 2009 IEEE.
Chen, P.Y., Van, L.D., Reddy, H.C. & Lin, C.T. 2009, 'A new VLSI 2-D fourfold-rotational-symmetry filter architecture design', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 93-96.
View/Download from: Publisher's site
In this paper, we propose two new two-dimensional (2-D) IIR and FIR filter architectures for 2-D transfer function using fourfold rotational symmetry. The presented type-I structure with fourfold rotational symmetry has the lowest number of multipliers, and zero latency. Importantly, the proposed type-II IIR filter possesses high speed, local broadcast, and the same number of multipliers and latency as the type I shows at expense of a slight increment of number of delay elements. ©2009 IEEE.
Huang, H.S., Pal, N.R., Ko, L.W. & Lin, C.T. 2009, 'Automatic identification of useful independent components with a view to removing artifacts from EEG signal', Proceedings of the International Joint Conference on Neural Networks, pp. 1267-1271.
View/Download from: Publisher's site
Removal of artifacts is an important step in any research in/application of electroencephalogram (EEG). The artifacts may contain eye-blinking, muscle noise, heart signal, line noise, and environmental effect. Such noises often make the raw EEG signals not very useful for extraction/identification of physiological phenomena from EEG. The independent component analysis (ICA) is a popular technique for artifact removal in brain research and some reports demonstrate that ICA can remove the artifacts with lower (acceptable) loss of information. But, these reports select useful independent components manually, primarily by looking at the scalp-plots. This is of great inconvenience and is a barrier for BCI or real-time applications of EEG. In this paper, we demonstrate that machine learning methods could be quite effective to discriminate useful independent components from artifacts and our findings suggests the possibility of developing a 'universal" machine for artifact removal in EEG. © 2009 IEEE.
Chen, P.Y., Van, L.-.D., Reddy, H.C. & Lin, C.T. 2008, 'A new VLSI 2-D diagonal-symmetry filter architecture design', IEEE Asia-Pacific Conference on Circuits and Systems, Proceedings, APCCAS, pp. 320-323.
View/Download from: Publisher's site
In this paper, we propose two new two-dimensional (2-D) IIR and FIR filter architectures for 2-D transfer function with diagonal symmetry. The presented type-I structure with diagonal symmetry has the lowest number of multipliers, and zero latency without sacrificing the number of the delay elements. Importantly, the proposed type-II IIR filter possesses high speed, local broadcast, and the same number of multipliers and latency as the type I shows at expense of a slight increment of number of delay elements. © 2008 IEEE.
Lin, C.T., Lin, H.Z., Chiu, T.W., Chao, C.F., Chen, Y.C., Liang, S.F. & Ko, L.W. 2008, 'Distraction-related EEG dynamics in virtual reality driving simulation', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 1088-1091.
View/Download from: Publisher's site
Driver distraction has been recognized as a significant cause of traffic incidents. Therefore, the aim of this study was to investigate Electroencephalography (EEG) dynamics in response to distraction during driving. To study human cognition under specific driving task, we used Virtual Reality (VR) based driving simulation to simulate events including unexpected car deviations and mathematics questions (math) in real driving. For further assessing effects of the stimulus onset asynchrony (SOA) between the deviation onset and math presented on the EEG dynamics, we designed five cases with different SOA. The scalp-recorded EEG channel signals were first separated into independent brain sources by Independent Component Analysis (ICA). Then, the Event-Related-Spectral-Perturbations (ERSP) measuring changes of EEG power spectra were used to evaluate the brain dynamics in time-frequency domains. Results showed that increases of theta band (5-7.8 Hz) and beta band (12.2-17 Hz) power were observed in the frontal cortex. Results demonstrated that reaction time and multiple cortical EEG sources responded to the driving deviations and math occurrences differentially in the stimulus onset asynchrony. Results also suggested that the theta band power increase in frontal area could be used as the distracted indexes for early detecting driver's inattention in the future. ©2008 IEEE.
Huang, W.C., Hung, S.H., Chung, J.F., Chang, M.H., Van, L.D. & Lin, C.T. 2008, 'FPGA implementation of 4-channel ICA for on-line EEG signal separation', 2008 IEEE-BIOCAS Biomedical Circuits and Systems Conference, BIOCAS 2008, pp. 65-68.
View/Download from: Publisher's site
Blind source separation of independent sources from their mixtures is a common problem for multi-sensor applications in real world, for example, speech or biomedical signal processing. This paper presents an independent component analysis (ICA) method with information maximiz ation (Infomax) update applied into 4-channel one-line EEG signal separation. This can be implemented on FPGA with a fixed-point number representation, and then the separated signals are transmitted via Bluetooth. As experimental results, the proposed design is faster 56 times than soft performance, and the correlation coefficients at least 80% with the absolute value are compared with off-line processing results. Finally, live demonstration is shown in the DE2 FPGA board, and the design is consisted of 16,605 logic elements. © 2008 IEEE.
Lin, C.T., Pal, N.R., Chuang, C.Y., Jung, T.P., Ko, L.W. & Liang, S.F. 2008, 'An EEG-based subject- and session-independent drowsiness detection', Proceedings of the International Joint Conference on Neural Networks, pp. 3448-3454.
View/Download from: Publisher's site
Monitoring and predicting human cognitive state and performance using physiological signals such as Electroencephalogram (EEG) have recently gained increasing attention in the fields of brain-computer interface and cognitive neuroscience. Most previous psychophysiological studies of cognitive changes have attempted to use the same model for all subjects. However, the relatively large individual variability in EEG dynamics relating to loss of alertness suggests that for many operators, group statistics cannot be used to accurately predict changes in cognitive states. Attempts have also been made to build a subject-dependent model for each individual based on his/her pilot data tb account for Individual variability. However, such methods assume the cross-session variability in EEG dynamics to be negligible, which could be problematic due to electrode displacements, environmental noises, and skin-electrode impedance. Here first we show that the EEG power in the alpha and theta bands are strongly correlated with changes in the subject's cognitive state reflected through his driving performance and hence his departure from alertness. Then under very mild and realistic assumptions we derive a model for the alert state of the person using EEG power in the alpha and theta bands. We demonstrate that deviations (computed by Mahalanobis distance) of the EEG power in the alpha and theta bands from the corresponding alert models are correlated to the changes in the driving performance. Finally, for detection of drowsiness we use a linear combination of deviations of the EEG power in the alpha band and theta band from the respective alert models that best correlates with subject's changing level of alertness, indexed by subject's behavioral response in the driving task. This approach could lead to a practical system for noninvasive monitoring of the cognitive state of human operators in attention-critical settings. © 2008 IEEE.
Chen, C.H., Liu, Y.C., Lin, C.J. & Lin, C.T. 2008, 'A hybrid of cooperative particle swarm optimization and cultural algorithm for neural fuzzy networks', IEEE International Conference on Fuzzy Systems, pp. 238-245.
View/Download from: Publisher's site
This study presents an evolutionary neural fuzzy network, designed using the functional-link-based neural fuzzy network (FLNFN) and a new evolutionary learning algorithm. This new evolutionary learning algorithm Is based on a hybrid of cooperative particle swarm optimization and cultural algorithm. It is thus called cultural cooperative particle swarm optimization (CCPSO). The proposed CCPSO method, which uses cooperative behavior among multiple swarms, can increase the global search capacity using the belief space. Cooperative behavior involves a collection of multiple swarms that interact by exchanging information to solve a problem. The belief space is the information repository in which the individuals can store their experiences such that other individuals can learn from them indirectly. The proposed FLNFN model uses functional link neural networks as the consequent part of the fuzzy rules. Finally, the proposed functional-link-based neural fuzzy network with cultural cooperative particle swarm optimization (FLNFN-CCPSO) is adopted in several predictive applications. Experimental results have demonstrated that the proposed CCPSO method performs well in predicting the time series problems. © 2008 IEEE.
Tsai, Y.S., Chung, I.F., Lin, C.T. & Pal, N.R. 2008, 'Identification of different sets of biomarkers for diagnostic classification of cancers', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 866-875.
View/Download from: Publisher's site
Accurate diagnosis of neuroblastoma, non-Hodgkin lymphoma, rhabdomyosarcoma, and Ewing sarcoma, is often difficult because these cancers appear similar in routine histology. Finding a few useful biomarkers (not all related genes) that can discriminate between the subgroups will help designing better diagnostic systems. In an earlier study we reported a set of seven genes having excellent discrimination power. In this investigation we extend that study and find other distinct sets of genes with strong class specific signatures. This is achieved analyzing the correlation between genes. This led us to find another set of seven genes with better discriminating power. Our original gene selection method used a neural network whose output may significantly depend on initialization of the network, network size as well as the training data set. To address these issues we propose a scheme based on re-sampling. This method can also reduce the effect wide variation in number of data points in the training set from different classes. This method led us to find a set of five genes with good discriminating power. The genes identified by the proposed methods have roles in cancer biology. © 2008 Springer-Verlag Berlin Heidelberg.
Lin, C.T., Liang, S.F., Chao, W.H., Ko, L.W., Chao, C.F., Chen, Y.C. & Huang, T.Y. 2007, 'Driving style classification by analyzing eeg responses to unexpected obstacle dodging tasks', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 4916-4919.
View/Download from: Publisher's site
Driving safely has received increasing attention of the publics due to the growing number of traffic accidents that the driver's driving style is highly correlated to many accidents. The purpose of this study is to investigate the relationship between driver's driving style and driver's ERP response. In our research, a virtual reality (VR) driving environment is developed to provide stimuli to subjects. Independent component analysis (ICA) is used to decompose the electroencephalogram (EEG) data. The power spectrum analysis of ICA components and correlation analysis are employed to investigate the EEG activities related to driving style. Experimental results demonstrate that we may classify the drivers into aggressive or gentle styles based on the observed ERP difference corresponding to the proposed unexpected obstacle dodging tasks. © 2006 IEEE.
Van, L.D., Lin, C.T. & Yu, Y.C. 2007, 'VLSI architecture for the low-computation cycle and power-efficient recursive DFT/IDFT design', IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, pp. 1644-1652.
View/Download from: Publisher's site
In this paper, we propose one low-computation cycle and power-efficient recursive discrete Fourier transform (DFT)/inverse DFT (IDFT) architecture adopting a hybrid of input strength reduction, the Chebyshev polynomial, and register-splitting schemes. Comparing with the existing recursive DFT/IDFT architectures, the proposed recursive architecture achieves a reduction in computation-cycle by half. Appling this novel low-computation cycle architecture, we could double the throughput rate and the channel density without increasing the operating frequency for the dual tone multi-frequency (DTMF) detector in the high channel density voice over packet (VoP) application. From the chip implementation results, the proposed architecture is capable of processing over 128 channels and each channel consumes 9.77 W under 1.2 V-20 MHz in TSMC 0.13 1P8M CMOS process. The proposed VLSI implementation shows the power-efficient advantage by the low-computation cycle architecture. © 2007 The Institute of Electronics, Information and Communication Engineers.
Lin, C.T., Chuang, S.W., Chen, Y.C., Ko, L.W., Liang, S.F. & Jung, T.P. 2007, 'EEG effects of motion sickness induced in a dynamic virtual reality environment', Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings, pp. 3872-3875.
View/Download from: Publisher's site
The Electroencephalogram (EEG) dynamics which relate to motion sickness are studied in this paper. Instead of providing visual or motion stimuli to the subjects to induce motion sickness, we employed a dynamic virtual-reality (VR) environment in our research. The environment consisted of a 3D surrounding VR scene and a motion platform providing a realistic situation. This environment provided the advantages of safety, low cost, and the realistic stimuli to induce motion sickness. The Motion Sickness Questionnaire (MSQ) was used to assess the sickness level, and the EEG effects on the subjects with high sickness levels were investigated using the independent component analysis (ICA). The fake-epoch extraction was then applied to the nausea-related independent components. Finally we employed the Event-Related Spectral Perturbation (ERSP) technology on the fake-epochs in order to determine the EEG dynamics during motion sickness. The experimental results show that most subjects experienced an 8-10 Hz power increase to their motion sickness-related phenomena in the parietal and motor areas. Moreover, some subjects experienced an EEG power increase of 18-20 Hz in their synchronized responses recorded in the same areas. The motion sickness-related effects and regions can be successfully obtained from our experimental results. ©2007 IEEE.
Wu, S.J., Wu, C.T., Chiou, Y.Y., Lin, C.T. & Chung, Y.N. 2007, 'Balancing control of sliding inverted-wedge system: Classical-method-based compensation', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 1349-1354.
View/Download from: Publisher's site
Root-locus-based PID controller and LQR-based controller always fail as system nonlinearity increases. We here propose the optimization-compensated block/diagram to reinforce the stabilization ability of these two classical control methods for nonlinear system, and besides, to achieve other performance requirements such as constrained overshoot and fast response. The controller design of a nonlinear sliding weights balancing mechanism is based on optimization-compensated root locus and LQR method. First, according to root-locus of the linearized dynamic system, we propose extra poles and zeros addition to roughly draw the locus shifting to left to achieve stabilization requirement. The poles and zeros are realized by P/PD/PID controllers. For LQR approach, we choose performance parameters to meet stabilization and minimum energy requirement. The controller is realized as feedback controller. Further, to compensate the model-error from nonlinearity and to meet other performance such as overshoot and setting time, some P/PID parameters for root-locus method and the feedback gain for LQR method are optimized via optimal parameter searching in NCD/Matlab toolbox. The simulation results demonstrate the stability and the constrained performances of the entire closed-loop system can be ensured by the proposed compensated control block diagrams. © 2006 IEEE.
Lin, C.T., Ko, L.W., Lin, Y.H., Jung, T.P., Liang, S.F. & Hsiao, L.S. 2007, 'EEG activities of dynamic stimulation in VR driving motion simulator', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 551-560.
The purpose of this study is to investigate Electroencephalography dynamics in response to kinesthetic stimuli during driving. We used a Virtual Reality driving simulator consisted of a hydraulic hexapod motion platform to create practical driving events. We compared the EEG dynamics in response to kinesthetic stimulus while the platform was in motion, to that while the platform was stationary. The scalp-recorded EEG channel signals were first separated into independent brain sources using Independent Component Analysis (ICA), and then studied with time-frequency analysis. Our results showed that independent brain processes near the somatomotor cortex exhibited alpha power decreases across sessions and subjects. Negative potentials phase-locked to the onsets of deviation events under motion conditions were observed in a central midline component. The results allow us to better understand different brain networks involved in driving, and provide a foundation for studying event-related EEG activities in the presence of kinesthetic stimuli. © Springer-Verlag Berlin Heidelberg 2007.
Lin, C.T., Hsieh, H.Y., Liang, S.F., Chen, Y.C. & Ko, L.W. 2007, 'Development of a wireless embedded brain - Computer interface and its application on drowsiness detection and warning', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 561-567.
The existing bio-signal monitoring systems are mostly designed for signal recording without the capability of automatic analysis so that their applications are limited. The goal of this paper is to develop a real-time wireless embedded electroencephalogram (EEG) monitoring system that includes multi-channel physiological acquisition, wireless transmission, and an embedded system. The wireless transmission can overcome the inconvenience of wire routing and the embedded multi-task scheduling for the dual-core processing system is developed to realize the real-time processing. The whole system has been applied to detect the driver's drowsiness for demonstration since drowsiness is considered as a serious cause of many traffic accidents. The electroencephalogram (EEG) features changes from wakefulness to drowsiness are extracted to detect the driver's drowsiness and an on-line warning feedback module is applied to avoid disasters caused by fatigue. © Springer-Verlag Berlin Heidelberg 2007.
Lin, C.T., Huang, Y.C., Mei, T.W., Pu, H.C. & Hong, C.T. 2007, 'Multi-objects tracking system using adaptive background reconstruction technique and its application to traffic parameters extraction', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2057-2062.
View/Download from: Publisher's site
In this paper, we present a real-time multi-objects tracking system which can detect various types of moving objects in image sequences of traffic video obtained from a stationary video camera. Using the adaptive background reconstruction technique can effectively handle with environmental changes and obtain good results of objects extraction. Besides, we introduce a robust region- and feature-based tracking algorithm with plentiful features to track correct objects continuously. After tracking objects successfully, we can analyze the tracked objects' properties and recognize their behavior for extracting some useful traffic parameters. According to the structure of our proposed algorithms, we implemented a tracking system including the functions of objects classification and accident prediction. Experiments were conducted on real-life traffic video of some intersection and testing datasets of other surveillance research. The results proved the algorithms we proposed achieved robust segmentation of moving objects and successful tracking with objects occlusion or splitting events. The implemented system also extracted useful traffic parameters. © 2006 IEEE.
Lin, C.T., Ko, L.W., Lin, K.L., Liang, S.F., Kuo, B.C., Chung, I.F. & Van, L.D. 2007, 'Classification of driver's cognitive responses using nonparametric single-trial EEG analysis', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 2019-2023.
Accidents caused by errors and failures in human performance among traffic fatalities have a high rate causing death and become an important issue in public security. The key problem causing these car accidents is mainly because that the drivers failed to perceive the changes of the traffic lights or the unexpected conditions happening accidentally on the roads. In this paper, we devised a quantitative analysis for ongoing assessment of driver's cognitive responses by investigating the neurobiological information underlying electroencephalographic (EEG) brain dynamics in traffic-light experiments in a virtual-reality (VR) dynamic driving environment. Three different feature extraction methods including Nonparametric Weighted Feature Extraction (NWFE), Principial Component Analysis (PCA), Discriminant Analysis Feature Extraction (DAFE) are applied to reduce the feature dimension and project the measured EEG signals to a feature space spanned by their eigenvectors. After that, the mapped data can be classified with fewer features and their classification results are compared by utilizing three different classifiers including Gaussian classifier (GC), k Nearest neighbor classification (KNNC), and Naive Bayes Classifier (NBC). Experimental results show that the successful rate of Nonparametric Weighted Feature Extraction combined with Gaussian classifier is higher more than 10% compared with other combinations. It also demonstrates the feasibility of detecting and analyzing single-trail ERP signals that represent operators' cognitive states and responses to task events. © 2007 IEEE. Electroencephalographic, Nonparametric Weighted Feature Extraction, Principial Component Analysis, Discriminant Analysis Feature Extraction, Gaussian classifier, k Nearest neighbor classification, Naive Bayes Classifier.
Lin, C.T., Chen, S.A., Cheng, Y.C. & Hong, C.T. 2007, 'CNN-based local motion estimation for image stabilization processing and its implementation', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 1816-1819.
View/Download from: Publisher's site
The objective of this paper is to investigate a novel design for local motion vectors (LMVs) of image sequences, which are often used in a digital image stabilization (IS) system. The IS technique removes unwanted shaking phenomenon in image sequences captured by hand-held camcorders. It includes two main parts such as motion estimation and compensation. Most of computation power occurs in the part of motion estimation. In order to reduce this complexity, an idea, which integrates an adaptive-threshold method and cellular neural networks (CNN) architecture, is designed to improve this problem. The design only implements the most important local motion estimation with the array size of 1925 pixels. Experimental results with HSPICE simulation and CNNLM are shown that the proposed architecture fast searches the location of possible LVMs and has the capability of real-time operations. ©2006 IEEE.
Hsieh, H.Y., Liang, S.F., Ko, L.W., Lin, M. & Lin, C.T. 2007, 'Development of a real-time wireless embedded brain signal acquisition/processing system and its application on driver's drowsiness estimation', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 4374-4379.
View/Download from: Publisher's site
In this paper, a portable real-time wireless embedded brain signal acquisition/processing system is developed. The proposed system integrates electroencephalogram signal amplifier technique, wireless transmission technique, and embedded real-time system. The development strategy of this system contains three parts: First, the Bluetooth protocol is used as a transmission interface and integrated with the bio-signal amplifier to transmit the measured physiological signals wirelessly. Second, the OMAP (Open Multimedia Architecture Platform) is used as a development platform and an embedded operating system for OMAP is also designed. Finally, DSP Gateway is developed as a mechanism to deal with the brain-signal analyzing tasks shared by ARM and DSP. A driver's cognitive-state estimation program has been developed and implemented on the proposed dual core processor-based real time wireless embedded system for demonstration. © 2006 IEEE.
Chen, C.H., Lin, C.T. & Lin, C.J. 2007, 'A functional-link-based fuzzy neural network for temperature control', Proceedings of the 2007 IEEE Symposium on Foundations of Computational Intelligence, FOCI 2007, pp. 53-58.
View/Download from: Publisher's site
This study presents a functional-link-based fuzzy neural network (FLFNN) structure for temperature control. The proposed FLFNN controller uses functional link neural networks (FLNN) that can generate a nonlinear combination of the input variables as the consequent part of the fuzzy rules. An online learning algorithm, which consists of structure learning and parameter learning, is also presented. The structure learning depends on the entropy measure to determine the number of fuzzy rules. The parameter learning, based on the gradient descent method, can adjust the shape of the membership function and the corresponding weights of the FLNN. Simulation result of temperature control has been given to illustrate the performance and effectiveness of the proposed model. © 2007 IEEE.
Chen, C.H., Lin, C.J. & Lin, C.T. 2007, 'A recurrent functional-link-based neural fuzzy system and its applications', Proceedings of the 2007 IEEE Symposium on Computational Intelligence in Image and Signal Processing, CIISP 2007, pp. 415-420.
View/Download from: Publisher's site
In this paper, a recurrent functional-link-based neural fuzzy system (RFLNFS) is proposed for prediction of time sequence and skin color detection. The proposed RFLNFS model uses functional link neural network as the consequent part of fuzzy rules. The RFLNFS model can generate the consequent part of a nonlinear combination of the input variables. The recurrent network is embedded in the RFLNFS by adding feedback connections in the second layer, where the feedback units act as memory elements. An online learning algorithm, which consists of structure learning and parameter learning, is also presented. Finally, the RFLNFS is applied to two simulations. The simulation results of the dynamic system modeling have shown that the RFLNFS model can solve the temporal problem and the RFLNFS model has superior performance than other models. © 2007 IEEE.
Lin, C.T., Hsieh, H.Y., Ko, L.W., Lin, M. & Liang, S.F. 2006, 'Development of portable wireless brain computer interface with embedded systems', IEEE 2006 Biomedical Circuits and Systems Conference Healthcare Technology, BioCAS 2006, pp. 226-229.
View/Download from: Publisher's site
In this paper, we propose an embedded multi-task scheduling system to make the wireless brain computer interface to real-time receive electroencephalogram signal more accurate. This method we propose also can increase the flexibility of the system. Base on the method, we can change the peripheral devices easily. And the tasks are taken as modules, so it also can increase the flexibility of system programs. The proposed system integrates electroencephalogram signal amplifier technique, wireless transmission technique, and embedded real-time system. The OMAP (Open Multimedia Architecture Platform) is used as a development platform and an embedded operating system is also used. © 2006 IEEE.
Lin, C.T., Liang, S.F., Chen, Y.C., Hsu, Y.C. & Ko, L.W. 2006, 'Driver's drowsiness estimation by combining EEG signal analysis and ICA-based fuzzy neural networks', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 2125-2128.
The public security has become an important issue in recent years, especially, the safe manipulation and control of vehicles in preventing the growing number of traffic accident fatalities. Accidents caused by drivers' drowsiness have a high fatality rate due to the decline of drivers' abilities in perception, recognition, and vehicle control abilities while sleepy. Preventing such an accident requires a technique for detecting, estimating, and predicting the level of alertness of a driver and a mechanism to maintain the driver's maximum performance of driving. The ICAFNN is a fuzzy neural network (FNN) capable of parameter self-adapting and structure selfconstructing to acquire a small number of fuzzy rules for interpreting the embedded knowledge of a system from the given training data set. Our experiments show that the ICAFNN can achieve significant improvements in the accuracy of drowsiness estimation compared with our previous works. © 2006 IEEE.
Lin, C.T. & Chin, C.L. 2006, 'Using fuzzy inference and cubic curve to detect and compensate backlight image', International Journal of Fuzzy Systems, pp. 2-13.
This paper proposes a new algorithm method for detection and compensation of backlight images. The proposed technique attacks the weaknesses of conventional backlight image processing methods, such as over-saturation and diminished contrast. This proposed algorithm consists of two operational phases, the detection phase and the compensation phase. In the detection phase, we use the spatial position characteristic and the histogram of backlight images to obtain two image indices which can determine the backlight degree of an image. The fuzzy inference method is then used to integrate these two indices into a final backlight index which determines the final backlight degree of an image more precisely. The compensation phase is used to solve the over-saturation problem which usually exists in conventional image compensation methods. In this phase, we propose the adaptive cubic curve method to compensate and enhance the brightness of backlight images. The luminance of a backlight image is adjusted according to the cubic curve equation which adapts dynamically according to the backlight degree indicated by the backlight index estimated in the detection phase. The performance of the proposed technique was tested against 300 backlight images covering a variety of backlight conditions and degrees. A comparison of the results of previous experiments clearly shows the superiority of our proposed technique in solving over-saturation and backlight detection problems. © 2006 TFSA.
Lin, C.T., Yu, A.C. & Van, L.D. 2006, 'A low-power 64-point FFT/IFFT design for IEEE 802.11a WLAN application', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 4523-4526.
In this paper, we propose a cost-effective and low-power 64-point fast Fourier transform (FFT)/inverse FFT (IFFT) architecture and chip adopting the retrenched 8-point FFT/IFFT (R8-FFT) unit and an efficient data-swapping method based output buffer unit The whole chip systematic performance concerning about the area, power, latency and pending cycles for the application of IEEE 802.11a WLAN standard has been analyzed. The proposed R8-FFT unit utilizing the symmetry property of the matrix decomposition achieves half computation-complexity and less power consumption compared with the recently proposed FFT/IFFT designs. On the other hand, applying the proposed data-swapping method, a low-cost and low-power output buffer can be obtained. So as to further increase system performance, we propose one scheme: the multiplication-afler-write (MAW) method. Applying MA W method with R8-FFT unit, the resulting FFT/IFFT design not only leads to the balancing pending cycle, but also abbreviating computation latency to 8 clock cycles. Consequently, adopting the above proposed two units and one scheme, the whole chip consumes 22.36mW under 1.2V@20 MHz in TSMC 0.13 1P8M CMOS process. © 2006 IEEE.
Lin, C.T., Chen, S.A., Cheng, Y.C. & Chung, J.F. 2006, 'CNN-based local motion estimation chip for image stabilization processing', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 2645-2648.
This paper is to investigate a novel design for local motion vectors (LMVs) of image sequences, which are often used in a digital image stabilization (IS) system. The IS technique removes unwanted shaking phenomenon in image sequences captured by hand-held camcorders. It includes two main parts such as motion estimation and compensation. Most of computation power occurs in the part of motion estimation. In order to reduce this complexity, an idea, which integrates an adaptive-threshold method and cellular neural networks (CNN) architecture, is designed to improve this problem. The design only implements the most important local motion estimation with the array size of 1925 pixels. Experimental results with HSPICE simulation and CNNUM are shown that the proposed architecture fast searches the location of possible LVMs and has the capability of real-time operations. The complete design has integrated into the total area of 8.1mm 2 by using TSMC 0.35m mixed-signal process. © 2006 IEEE.
Lu, S.M., Liang, S.F. & Lin, C.T. 2006, 'A HVS-directed neural-network-based approach for salt-pepper impulse noise removal', Journal of Information Science and Engineering, pp. 925-939.
In this paper, a novel two-stage noise removal algorithm to deal with salt-pepper impulse noise is proposed. In the first stage, the decision-based recursive adaptive noise-exclusive median filter is applied to remove the noise cleanly and to keep the uncorrupted information as well as possible. In the second stage, the fuzzy decision rules inspired by human visual system (HVS) are proposed to classify image pixels into human perception sensitive class and non-sensitive class. A neural network is proposed to compensate the sensitive regions for image quality enhancement. According to the experimental results, the proposed method is superior to conventional methods in perceptual image quality as well as the clarity and the smoothness in edge regions of the resultant images.
Huang, C.H. & Lin, C.T. 2006, 'Image enhancement algorithm for hexagonal cellular neural networks', IEEE Asia-Pacific Conference on Circuits and Systems, Proceedings, APCCAS, pp. 386-389.
View/Download from: Publisher's site
In this paper, we propose an image enhancement algorithm for a display. Traditionally, high dynamic range algorithms handle the conversion from the scene in the real world to the screen in a display since the optical-physical conditions are changed. Similar reasons motivated our study of analyzing the relationship between the display and the Human Vision System (HVS). In this paper, we introduce an image enhancement algorithm, which is based on a well-known high-dynamic range compression algorithm, named Retinex theory. Retinex theory provides an approach of separating the illumination from the reflectance in a given image and thereby compensating for nonuniform lighting. The proposing algorithm is implemented on an advance Cellular Neural Network structure, the hexagonal-type Cellular Neural Network (hCNN). Via examining the stable central linear system of a hCNN, we are able to implement the Retinex theory and operate the CNN in the stable region. Meanwhile, we propose an approach to estimate the parameters in the Retinex theory based on the analysis of the interactions between the retina and the display. Those parameters vary depending on the environment and usually are difficult to obtain. Proposing algorithm is based on biological inspired technology. In our experiments, some quite good results are obtained. ©2006 IEEE.
Huang, C.H., Koeppl, H. & Lin, C.T. 2006, 'A bio-inspired computer fovea model based on hexagonal-type cellular neural networks', IEEE International Conference on Neural Networks - Conference Proceedings, pp. 5189-5195.
In this work we propose a novel computer fovea model based on hexagonal-type Cellular Neural Networks (hCNN). The hCNN represents a new image processing architecture that is motivated by the overwhelming evidence for hexagonal image processing in biological systems. The necessary new coupling templates and basic hCNN image operators are introduced. The fovea model includes the biological mechanisms of the photoreceptors, the horizontal cells, the ganglions, the bipolar cells, and their cooperation. Thus the model describes the signal processing from the optical stimulation at retina to the output of the ganglion cells. Different building blocks of the model turned out to be useful for practical image enhancement algorithms. Two such applications are considered in this work, namely the image sharpness improvement and the color constancy algorithm. © 2006 IEEE.
Chiou, J.C., Ko, L.W., Lin, C.T., Hong, C.T., Jung, T.P., Liang, S.F. & Jeng, J.L. 2006, 'Using novel MEMS EEG sensors in detecting drowsiness application', IEEE 2006 Biomedical Circuits and Systems Conference Healthcare Technology, BioCAS 2006, pp. 33-36.
View/Download from: Publisher's site
Electroencephalographic (EEG) analysis has been widely adopted for the monitoring of cognitive state changes and sleep stages because abundant information in EEG recording reflects changes in drowsiness, arousal, sleep, and attention, etc. In this study, Micro-Electro-Mechanical Systems (MEMS) based silicon spiked electrode array, namely dry electrodes, are fabricated and characterized to bring EEG monitoring to the operational workplaces without requiring conductive paste or scalp preparation. An isotropic/anisotropic reactive ion etching with inductive coupled plasma (RIE-ICP) micromachining fabrication process was developed to manufacture the needle-like micro probes to pierce the stratum corneum of skin and obtain superior electrically conducting characteristics. This article reports a series of prosperity testing and evaluations of continuous EEG recordings. Our results suggest that the dry electrodes have advantages in electrode-skin interface impedance, signal intensity and size over the conventional (wet) electrodes. In addition, we also developed an EEG-based drowsiness estimation system that consists of the dry-electrode array, power spectrum estimation, Principal Component Analysis (PCA)-based EEG signal analysis, and multivariate linear regression to estimate driver's drowsiness level in a virtual-reality-based dynamic driving simulator to demonstrate the potential applications of the MEMS electrodes in operational environments. ©2006 IEEE.
Lin, C.T., Chen, Y.C., Wu, R.C., Liang, S.F. & Huang, T.Y. 2005, 'Assessment of Driver's Driving Performance and Alertness Using EEG-based Fuzzy Neural Networks', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 152-155.
View/Download from: Publisher's site
Traffic fatalities in recent years have become a serious concern to our society. Accidents caused by drivers' drowsiness have a high fatality rate due to the decline of drivers' abilities in perception, recognition, and vehicle control abilities while sleepy. Preventing such an accident requires a technique for detecting, estimating, and predicting the level of alertness of a driver and a mechanism to maintain the driver's maximum performance of driving. This paper proposed a system that combines electroencephalogram (EEG) power spectra estimation, independent component analysis and fuzzy neural network models to estimate drivers' cognitive state in a dynamic virtual-reality-based driving environment. Experimental results show that the quantitative driving performance can be accurately and successfully estimated through analyzing driver's EEG signals by the proposed system.©2005 IEEE.
Lin, C.T., Liang, S.F., Yeh, C.M. & Fan, K.W. 2005, 'Fuzzy neural network design using support vector regression for function approximation with outliers', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2763-2768.
A fuzzy neural network based on support vector learning mechanism for function approximation is proposed in this paper. Support vector regression (SVR) is a novel method for tackling the problems of function approximation and regression estimation based on the statistical learning theory. SVR has been shown to have robust properties against noise. A novel support-vector-regression based fuzzy neural network (SVRFNN) by integrating SVR technology into FNN is developed. The SVRFNN combines the high accuracy and robustness of support vector regression (SVR) and the efficient human-like reasoning of FNN for function approximation. Experimental results show that the proposed SVFNN for function approximation can achieve good approximation performance with drastically reduced number of fuzzy kernel functions. © 2005 IEEE.
Lin, C.T., Chao, W.H., Chen, Y.C. & Liang, S.F. 2005, 'Adaptive feature extractions in an EEG-based alertness estimation system', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2096-2101.
During the past years, the public security has become an important issue, especially, the safe manipulation and control of various vehicles. Maintaining high cognition is particularly important for the drivers behind the steering wheel. It requires an optimal estimation system to online continuously detect drivers' cognitive state related to abilities in perception, recognition, and vehicle control. In this paper, we proposed an EEG-based alertness estimation system with automatic feature selection mechanism. The independent component analysis (ICA) is used first to decompose the measured electroencephalogram (EEG). Then, a time-frequency analysis is performed to evaluate the time-frequency characteristic of each ICA component. We also proposed a new adaptive feature extracting mechanism for selections of frequency bands and ICA components. Different ranges of the alpha rhythm of subjects can be evaluated by the adaptive feature extracting mechanism according to the correlation coefficient between the ICA time-frequency response and the driving performance. The extracted features are then trained both by linear regression model and Self-cOnstructing Neuro-Fuzzy Inference Network (SONFIN) for the estimation of driving performance. The training and testing results of SONFIN are 96% and 91%, while the results of linear regression model are 90% and 85%, respectively. It demonstrates that the proposed adaptive feature extracting mechanism can achieve a great performance in alertness estimation with frequency band and component selection. © 2005 IEEE.
Lin, C.T., Fan, K.W. & Cheng, W.C. 2005, 'An illumination estimation scheme for color constancy based on chromaticity histogram and neural network', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2488-2494.
This paper proposes an algorithm for estimation the illumination of an image for the purpose of color constancy using chromaticity histogram. This method evaluates the color temperature of the light source by detecting the distribution of chromaticity histogram in an image. It has the advantages of high efficiency, good robustness, and no strict assumptions. We subsequently propose to use a multilayer perceptrons trained by Backpropagation algorithm to model the nonlinear functional relationship between the chromaticity histogram and coefficients of illuminant functions. The trained neural network can then be used to estimate the spectral power distribution of light source. Finally, we use the trained neural network to estimate spectral power distribution in a finite-dimensional linear model of surface reflectance for color constancy. For performance evaluation, two color-recovery experiments on synthetic images and nature images captured from a still digital camera are performed in this paper. All the results are compared to those of two existing popular algorithms (Max-RGB and Gray-World algorithms). © 2005 IEEE.
Lin, C.T. & Shou, Y.W. 2005, 'Texture classification and representation by CNN-based feature extraction', Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications, pp. 210-213.
This paper proposes a novel approach for texture classification by feature extraction based on cellular neural networks (CNN's) and an intelligent arrangement in the design or exact using of the templates in CNN's. This paper also gives a two-way processing mechanism including the analysis of features extracted from the output of CNN's mapping and a selective training step for obtaining the specific templates in CNN's by genetic algorithms (GA's) for more complicated texture patterns. In this paper, we introduce a one dimensional feature curve as well to indicate the characteristics of original texture patterns from the mapping of the output of CNN's for the latter texture classification. The method introduced in this paper could adaptively choose an appropriate processing procedure for the specific issues towards in specific texture patterns. We finally divide our experiments into two sections, for simple and advanced problems in texture classification. Our experimental results demonstrate the valid template training and at the same time show a satisfactory classification outcome in both defined texture problems.
Lin, C.T. & Tsai, T.H. 2005, 'Biological-inspired model for hybrid-order chromatic texture segregation', Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications, pp. 214-218.
In this study, a computational model for chromatic texture segregation is developed. The model attempts to simulate the visual processing characteristics by mimicking the visual perception. According to Hering's opponent theory, we deal with color information in a color space with three opponent axes. The algorithm extracts 1st-order features by a Gaussian filter and 2 nd-order features by a set of Gabor filters as so called Gabor wavelets. The hybrid-order features are combined at a common site to detect the boundary. The model is very intuitive and physiological relevant such that it reserves opportunities for further approaches.
Chen, C.H. & Lin, C.T. 2005, 'Identification of chaotic system using recurrent compensatory neuro-fuzzy systems', Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications, pp. 15-18.
In this paper, a Recurrent Compensatory Neuro-Fuzzy System (RCNFS) is proposed for identification and prediction. The compensatory-based fuzzy reasoning method is using adaptive fuzzy operations of neuro-fuzzy systems that can make the fuzzy logic systems more adaptive and effective. The recurrent network is embedded in the RCNFS by adding feedback connections in the second layer, where the feedback units act as memory elements. Also, an on-line learning algorithm is proposed to automatically construct the RCNFS. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter learning. Finally, the RCNFS is applied in several simulations. The simulation results of the dynamic system modeling have shown that 1) the RCNFS model converges quickly; 2) the RCNFS model requires a small number of tuning parameters; 3) the RCNFS model can solve the temporal problems and approximate a dynamic system.
Lin, C.T. & Huang, C.H. 2005, 'A complex texture classification algorithm based on gabor-type filtering cellular neural networks and self-organized fuzzy inference neural networks', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 3942-3945.
View/Download from: Publisher's site
In this paper, a bio-inspired complex texture classification algorithm has been introduced. This algorithm included two neural networks; One is Gabor-type Filtering Cellular Neural Networks which used to simulate human retina, the other is Self-Organized Fuzzy Inference Neural Networks which used to simulate the brain. Both the neural networks can be considered as feed-forward neural networks. Thus, we can say that the whole system which has been introduced in this paper is also a feed-forward system and which contains the ability of parallel processing. © 2005 IEEE.
Lin, C.Y., Lin, K.L., Huang, C.D., Chang, H.M., Yang, C.Y., Lin, C.T., Tang, C.Y. & Hsu, D.F. 2005, 'Feature selection and combination criteria for improving predictive accuracy in protein structure classification', Proceedings - BIBE 2005: 5th IEEE Symposium on Bioinformatics and Bioengineering, pp. 311-315.
View/Download from: Publisher's site
The classification of protein structures is essential for their function determination in bioinformatics. The success of the protein structure classification depends on two factors: the computational methods used and the features selected. In this paper, we use a combinatorial fusion analysis technique to facilitate feature selection and combination for improving predictive accuracy in protein structure classification. When applying these criteria to our previous work, the resulting classification has an overall prediction accuracy rate of 87% for four classes and 69.6% for 27 folding categories. These rates are significantly higher than our previous work and demonstrate that combinatorial fusion is a valuable method for protein structure classification. © 2005 IEEE.
Cheng, W.C. & Lin, C.T. 2005, 'A novel post-nonlinear ICA-based reflectance model for 3D surface reconstruction', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 3023-3026.
View/Download from: Publisher's site
We propose a novel reflectance model for photometric stereo. It consists of the diffuse components and the specular components. Unlike the past methods, we do not need to separate the two components from the nonlinear reflection model. We use a unsupervised learning adaptation algorithm to estimate the reflectance model based on image intensities. At first, the technique of the post nonlinear independent components analysis (ICA) model is used to obtain the surface normal on each point of an image. Then, the 3D surface model can be reconstructed based on the estimated surface normal on each point of image by using the method of enforcing integrability. We test our algorithm on synthetically generated images for the reconstruction of surface of objects and on a number of real images captured from the Yale Face Database B. The results clearly indicate the superiority of the proposed nonlinear reflectance model over the Georghiades's approach and the Hayakawa's approach. © 2005 IEEE.
Liang, S.F., Lin, C.T., Wu, R.C., Huang, T.Y. & Chao, W.H. 2005, 'Classification of Driver's Cognitive Responses From EEG Analysis', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 156-159.
View/Download from: Publisher's site
the past years, the growing number of traffic fatalities has become an important issue in public security. In this paper, we develop a quantitative analysis for ongoing assessment of cognitive response by investigating the neurobiological brain dynamics in traffic-light experiments. A single-trial event-related-potential (ERP)-based fuzzy neural network (FNN) is applied to recognize different brain potentials stimulated by red/green/yellow traffic-light events. The system consists of a dynamic virtual-reality (VR)-based motion simulation platform, EEG signal detection and analysis units, and FNN-based classifier. The independent component analysis (ICA) algorithms are used to obtain noise-free ERP signals from the multi-channel EEG signals. A novel temporal filter is also proposed to solve time-alignment problems of ERP features and principle component analysis (PCA) is used to reduce dimension of features, which were then fed into a FNN classifier. Experimental results demonstrate the feasibility of detecting and analyzing multiple streams of ERP signals that organize operators' cognitive responses to task events. Comparisons of three kinds of linear and nonlinear classifiers show that our proposed FNN-based classifier can achieve a satisfactory and superior recognition rate (85%). The classification results can be further transformed as the control/biofeedback signals of intelligent driving systems.©2005 IEEE.
Chung, J.F., Liu, D.J. & Lin, C.T. 2005, 'Multiband room effect simulator for 5.1-channel sound system', Proceedings - IEEE International Symposium on Circuits and Systems, pp. 2859-2862.
View/Download from: Publisher's site
This paper investigates a new room effect simulator with different quality of audio sound sources for 5.1-channel sound system. The audio effect relates its impulse response with different room-dimension. The multiband equalizer is designed to generate many kinds of music impression. We also modify the mono Dolby surround decoder into a simplified surround decoder so that the stereo room simulator can generate 5.1-channel surround outputs. All experiments are at a sampling rate of 44.1k Hz and 16 bits per sample. By the room effect simulator, we can get a full and a live listening experience. © 2005 IEEE.
Lin, C.T. & Huang, C.H. 2005, 'Cellular neural networks for Hexagonal Image Processing', Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications, pp. 81-84.
In this paper, Cellular Neural Networks (CNNs) for Hexagonal Image Processing (HIP) frameworks has been proposed. In this paper, we combine two distinguished researches; one is CNNs, which provide efficient computing abilities. The other is HIP, which contains most compact structure. Both CNNs and HIP are inspired by biological. CNNs present the behaviors which are most similar to the retina of human's eyes, and HIP presents architecture which is also most similar to the distribution of cells on the retina.
Cheng, Y.C., Chung, J.F., Lin, C.T. & Hsu, S.C. 2005, 'Local motion estimation based on cellular neural network technology for image stabilization processing', Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications, pp. 286-289.
This paper presents a novel robust image stabilization (IS) technique to find out local motion vectors in the image sequences captured. Our technique is based on a Cellular Neural Network (CNN) algorithm, which tracks a small set of features to estimate the motion of the camera. Real-time and parallel analog computing elements are contained in the architecture of CNN. It is a regular two-dimensional array and connects with its neighborhood locally. To implement this algorithm on VLSI CNN, the adaptive-minimized threshold method is proposed to find quickly extract reliable motion vectors in plain images which are lack of features or contain large low-contrast area. Each size of CNN is set to 1/120 of an image. A background evaluation model is also developed to deal with irregular images which contain large moving objects. The experimental results are on-line available to demonstrate the remarkable performance of the proposed CNN-based motion technique.
Lin, C.T. & Chen, S.A. 2005, 'Biological visual processing for hybrid-order texture boundary detection with CNN-UM', Proceedings of the IEEE International Workshop on Cellular Neural Networks and their Applications, pp. 146-149.
This paper investigates a novel biological visual processing for hybrid-order texture boundary detection. The texture boundary detection is based on the first- and second-order features to model the pre-attentive stage of a human visual system. This system is implemented by a cellular neural network universal machine (CNN-UM) with 33 templates to approximate desired filter transfer functions. The system design can process a 6464 gray-scale image. The proposed algorithm can successfully be performed by CNN-UM and detect the texture boundary in a given image.
Lin, C.T. & Huang, C.H. 2004, 'Texture boundary detection based on multiple and parrllel cellular neural networks', 11th IEEE International Conference on Electronics, Circuits and Systems, ICECS 2004, pp. 222-225.
In this paper, a texture boundary detection model has been introduced. It is inspired by the architecture of retina - the film of human's eyes. Based on the innovation - a multiple and parallel CNN processors, we implemented a retina-like model and thus the boundary of texture can be obtained. ©2004 IEEE.
Chen, S.A., Chung, J.F., Liang, S.F. & Lin, C.T. 2004, 'Cellular Neural Network (CNN) circuit design for modeling of early-stage human visual system', 2004 IEEE International Workshop on Biomedical Circuits and Systems.
This paper proposes a novel CNN-based biological visual processing for hybrid-order texture boundary detection. The texture boundary detection is based on the first- and second-order features to model pre-attentive stage of human visual system. This system is implemented by using a parallel computing neural network, called cellular neural networks (CNN). This CNN design adopts the multi-layer architecture involving a 55 large neighborhood and is extended to be the 1616 array size for image processing. The proposed circuit models have been verified and the proposed method can successfully detect the texture boundary in an image. © 2004 IEEE.
Lin, C.T., Chen, S.A., Huang, C.H. & Chung, J.F. 2004, 'Cellular Neural Networks and PCA neural networks based rotation / scale invariant texture classification', IEEE International Conference on Neural Networks - Conference Proceedings, pp. 153-158.
In this paper, we proposed a new index which can be used to classify the texture image. Because of the adjustment of image capture device or the distortion of image capture, the texture image may be transformed. Usually those transformations included rotation and scale. The proposed method provides an algorithm to avoid those effects respectively. This approach is the combination of Cellular Neural Networks and Principle Component Analysis Neural Networks. This fact implies it is a feed-forward neural networks, and it does not need any training set.
Hsu, C.F. & Lin, C.T. 2004, 'New techniques for intelligent control', IEEE International Symposium on Intelligent Control - Proceedings, pp. 13-18.
This paper proposed a new technique for the intelligent control approach, which is termed as fuzzy-identification-based adaptive fuzzy control (FIAFC) scheme. The developed FIAFC is compared of a principal controller and a robust controller. The principal controller utilizes an adaptive fuzzy model to identify the dynamics of the controlled system. The robust controller is designed to dispel the model error introduced by the adaptive fuzzy model. In the conventional adaptive fuzzy control (AFC) system, the adaptive law was designed to drive the tracking error to zero without considering the modeling error. In the proposed FIAFC system, not only the tracking-error information but also the modeling-error information are utilized in the derived adaptive law, thus the convergence performance can be improved. To investigate the effectiveness of the proposed FIAFC, it is applied to a chaotic system control. The simulation results have demonstrated that the proposed FIAFC can achieve better tracking performance than the AFC. Furthermore, the improvements are achieved at a negligible increase in the computational complexity. © 2004 IEEE.
Wu, R.C., Liang, S.F., Lin, C.T. & Hsu, C.F. 2004, 'Applications of event-related-potential-based brain computer interface to intelligent transportation systems', Conference Proceeding - IEEE International Conference on Networking, Sensing and Control, pp. 813-818.
In this paper, an event-related-potential (ERP) -based brain computer interface (BCI) is proposed for the application of intelligent transportation systems (ITS). It consists of a virtual-reality (VR) motion simulation platform and an electroencephalographic signal detection and analysis system. The goals are to demonstrate the feasibility of detecting and analyzing multiple streams of ERP signals that organize operators' cognitive states and responses to task events, and to develop an ERP-based brain computer interface to meet the requirements of public security of intelligent transportation systems. We setup detailed experimental procedures to perform the cognitive tasks and collect high-fidelity ERP signals in the well-controlled VR-based laboratory environments. The independent component analysis (ICA) algorithms are applied to separate and get noise-free ERP signals from the multi-channel measured signals. Experimental results show that the separated ERP signals achieve a satisfactory result and can be further classified and transformed as the control/monitoring signals of safety-driving system for ITS.
Wu, R.C., Lin, C.T., Liang, S.F., Huang, T.Y. & Jung, T.P. 2004, 'EEG-based fuzzy neural network estimator for driving performance', Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 4034-4040.
View/Download from: Publisher's site
Accidents caused by drivers' drowsiness have a high fatality rate because of the marked decline in the drivers' vehicle control abilities. Preventing accidents caused by drowsiness is highly desirable but requires techniques for continuously detecting, estimating, and predicting the level of alertness of drivers. This paper proposes a brain-machine interface that combines electroencephalographic power spectrum estimation, principal component analysis, and fuzzy neural networks to estimate/predict drivers' drowsiness level in a virtual-reality-based driving simulator. The driving performance is defined as deviation between the center of the vehicle and the center of the cruising lane. Our results demonstrated that the proposed method is feasible to accurately estimate quantitatively driving performance in a realistic driving simulator. © 2004 IEEE.
Lin, C.T., Yeh, C.M. & Hsu, C.F. 2004, 'Fuzzy neural network classification design using support vector machine', Proceedings - IEEE International Symposium on Circuits and Systems.
Fuzzy neural networks (FNNs) for pattern classification usually use the backpropagation or C-cluster type learning algorithms to learn the parameters of the fuzzy rules and membership functions from the training data. However, such kinds of learning algorithms usually cannot minimize the empirical risk (training error) and expected risk (testing error) simultaneously, and thus cannot reach a good classification performance in the testing phase. To tackle this drawback, a support-vector-based fuzzy neural network classification (SVFNNC) is proposed. The SVFNNC combines the superior classification power of support vector machine (SVM) in high dimensional data spaces and the efficient human-like reasoning of FNN in handling uncertainty information. The learning algorithm consists of two learning phases. In the phase 1, the fuzzy rules and membership functions are automatically determined by the clustering principle. In the phase 2, the parameters of FNN are calculated by the SVM with the proposed adaptive fuzzy kernel function. To investigate the effectiveness of the proposed SVFNNC, it is applied to the Iris, Vehicle and Dna datasets. Experimental results show that the proposed SVFNNC can achieve good classification performance with drastically reduced number of fuzzy kernel functions.
Liu, S.H., Lin, C.T., Wen, Z.C. & Wang, J.J. 2004, 'Using the system identify theorem for cnstructing the dynamic compliance of the brachial artery', 2004 IEEE International Workshop on Biomedical Circuits and Systems.
A noninvasive measurement technique with oscillometry, system identify, and the related measured circuits is investigated to detect the dynamic compliance of brachial artery. In oscillometry, oscillation amplitudes (OAs) embedded in the cuff pressure are effected by the arterial characteristic, body tissue, and cuff characteristic. In cuff deflation, pressure transducer and micro flower meter were used to detect the variation of cuff pressure and volume. A system identify theorem was used to reconstruct the cuff model. Using the cuff pressure and OAs, the arterial volume change was calculated under the different transmural pressure. This measurement system also detected the systolic and diastolic pressure, simultaneously. Therefore, the dynamic Pressure-Volume (P-V) curve of artery was made. © 02004 IEEE.
Lin, C.T. & Cheng, W.C. 2004, 'An on-line ICA-mixture-model-based fuzzy neural network', IEEE International Conference on Neural Networks - Conference Proceedings, pp. 2141-2146.
View/Download from: Publisher's site
This paper proposes a new fuzzy neural network (FNN) capable of parameter self-adapting and structure self-constructing to acquire a small number of fuzzy rules for interpreting the embedded knowledge of a system from the given training data set. The proposed FNN is inherently a modified Takagi-Sugeno-Kang (TSK)-type fuzzy rule-based model with neural network's learning ability. There are no rules initiated at the beginning and they are created and adapted through the newly proposed on-line independent component analysis (ICA) mixture model and back-propagation algorithm learning processing that performs simultaneous structure and parameter identification. Several experiments covering the areas of system identification and classification are carried out. These results show that the proposed FNN can achieve significant improvements in the convergence speed and prediction accuracy with simpler network structure.
Wu, R.C., Lin, C.T., Liang, S.F., Huang, T.Y., Chen, Y.C. & Jung, T.P. 2004, 'Estimating driving performance based on EEG spectrum and fuzzy neural network', IEEE International Conference on Neural Networks - Conference Proceedings, pp. 585-590.
The growing number of traffic fatalities in recent years has become a serious concern to society. Accidents caused by drivers' drowsiness behind the steering wheel have a high fatality rate because of the marked decline in the drivers' abilities of perception, recognition, and vehicle control abilities while sleepy. Preventing accidents caused by drowsiness requires a technique for detecting, estimating, and predicting the level of alertness of a driver and a mechanism for maintaining his/her maximum performance. This paper describes a system that combines electroencephalographic (EEG) power spectrum estimation, principal component analysis, and fuzzy neural network model to estimate/predict drivers' drowsiness level in a driving simulator. Our results demonstrated that, for the first time, it is feasible to accurately estimate task performance, accurately estimate quantitatively measured driving performance, expressed as deviation between the center of the vehicle and the center of the cruising lane, in a realistic driving simulation.
Lu, S.M., Pu, H.C. & Lin, C.T. 2003, 'A HVS-directed neural-network-based approach for impulse-noise removal from highly corrupted images', Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, pp. 72-77.
In this paper, a novel two-stage noise removal algorithm to deal with fixed-value impulse noise is proposed. In the first stage, the decision-based recursive adaptive median filter is applied to remove the noise cleanly and keep the uncorrupted information as well as possible. In the second stage, the fuzzy decision rules inspired by human visual system (HVS) are proposed to classify pixels of the image into human perception sensitive class and non-sensitive class. A neural network is proposed to enhance the sensitive regions to perform better visual quality. According to the experiment results, the proposed method is superior to conventional methods in perceptual image quality as well as the clarity and the smoothness in edge regions.
Pu, H.C., Lin, C.T., Liang, S.F. & Kumar, N. 2003, 'A novel neural-network-based image resolution enhancement', IEEE International Conference on Fuzzy Systems, pp. 1428-1433.
In this paper, a novel HVS-directed neural-network-based adaptive interpolation scheme for natural image is proposed. A fuzzy decision system built from the characteristics of the human visual system (HVS) is proposed to classify pixels of the input image into human perception non-sensitive class and sensitive class. High-resolution digital images along with supervised learning algorithms are used to automatically train the proposed neural network. Simulation results demonstrate that the proposed new resolution enhancement algorithm can produce higher visual quality of the interpolated image than the conventional interpolation methods.
Lin, C.T., Lee, T.T., Hsu, C.F. & Lin, C.M. 2003, 'Hybrid Adaptive Fuzzy Control Wing Rock Motion System with H Robust Performance', Proceedings of the International Joint Conference on Neural Networks, pp. 2372-2377.
In this paper, a hybrid adaptive fuzzy control (HAFC) system is developed for a wing rock motion system. The design of HAFC system contains three parts: one is an indirect controller, the other is a direct controller and the last is a robust controller. A weighting factor , which can be adjusted by a tradeoff between plant knowledge and control knowledge, is adopted to sum together the control efforts from the indirect and direct controllers. The robust controller is designed to achieve favorable control performance with a desired robustness. Simulation results demonstrate that the HAFC system can achieve favorable desired tracking performances for unknown the wing rock motion dynamics.
Duh, F.B., Juang, C.F. & Lin, C.T. 2003, 'Application of neural fuzzy network to pulse compression with binary phase code', IEEE International Conference on Fuzzy Systems, pp. 1389-1394.
To solve the existing dilemma between making good range resolution and maintaining the low average transmitted power, it is necessary for the pulse compression processing to give low range sidelobes in the modern high-resolution radar systems. The traditional pulse compression algorithms based on 13-element Barker code such as direct autocorrelation filter (ACF), least squares (LS) inverse filter, and linear programming (LP) filter have been developed, and the neural network algorithms were issued recently. However, the traditional algorithms cannot achieve the requirement of high signal-to-sidelobe ratio, and the normal neural network such as backpropagation (BP) network usually produces the extra problems of low convergence speed and sensitive to the Doppler frequency shift. To overcome these defects, a new approach using a neural fuzzy network with binary phase code to deal with pulse compression in a radar system is presented in this paper. The 13-element Barker code used as the binary phase signal code is carried out by six-layer self-constructing neural fuzzy network (SONFIN) with supervised learning algorithm. Simulation results show that this neural fuzzy network pulse compression (NFNPC) algorithm has the significant advantages in noise rejection performance, range resolution ability and Doppler tolerance, which are superior to the traditional and BP algorithms, and has faster convergence speed than BP algorithm.
Lin, C.T., Liu, D.J., Wu, R.C. & Wu, G.D. 2002, 'Noisy speech segmentation/enhancement with multiband analysis and neural fuzzy networks', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 301-309.
© Springer-Verlag Berlin Heidelberg 2002.Background noise added to speech can decrease the performance of speech segmentation and enhancement. To solve this problem, new methods have been developed in this thesis. First, a new speech segmentation method (ATF-based SONFIN algorithm) is proposed in fixed noise-level environment. This method contains the multiband analysis and a neural fuzzy network, and it achieves higher recognition rate than the TF-based robust algorithm by 5%. In addition, a new speech segmentation method called RTF-based RSONFIN algorithm is proposed for variable noise-level environment. The RTF-based RSONFIN algorithm contains a recurrent neural fuzzy network. This method contains the multiband analysis and achieve higher recognition rate than the TF-based robust algorithm by 12%.
Liu, S.H., Chung, I.F. & Lin, C.T. 2002, 'An optimal controller with synthetic fuzzy logic for tracking mean arterial pressure', IEEE International Conference on Fuzzy Systems, pp. 402-407.
This paper proposes a new noninvasive measurement method for tracking the tendency of mean arterial pressure (MAP) in the radial artery. The designed system consists of a tonometer, a micro syringe device, and a model-based fuzzy logic controller. The modified flexible diaphragm tonometer is to detect the continuous blood pressure waveform and vessel volume pulse. A precise mathematical model describing the interaction between the tonometer and artery is derived. To reach accurate measurement without distortion, an optimal controller is designed to compensate the change of MAP by applying a counter pressure on the tonometer chamber through the micro syringe device. Simulation results show that, for the real physiologic MAP with changing rates up to 20 or -20 mm-Hg/minute, the optimal controller can beat-to-beat adjust the tonometer's chamber pressure to follow the tendency of MAP accurately.
Duh, F.B. & Lin, C.T. 2002, 'Radar tracking for a maneuvering target using neural fuzzy based Kalman filter', IEEE International Conference on Fuzzy Systems, pp. 1405-1409.
A fast target maneuver detecting and highly accurate tracking technique using a neural fuzzy network based on Kalman filter is proposed in this paper. In the automatic target tracking system, there exists an important and difficult problem: how to detect the target maneuvers and fast response to avoid miss-tracking? To solve this problem, neural network and fuzzy algorithms have been issued recently. However, the normal neural networks such as backpropagation networks usually produce the extra problems of low convergence speed and/or large network size, and the fuzzy algorithms are not easy to partition the parameters. To overcome these defects and to make use of neural learning ability, a developed standard Kalman filter with a self-constructing neural fuzzy inference network (KF-SONFIN) algorithm for target tracking is presented in this paper. Without having to change the structure of Kalman filter nor modeling the maneuvering target, SONFIN algorithm, can always find itself an economic network size with a fast learning process. Simulation results show that the KF-SONFIN is superior to the traditional IE and VDF methods in estimation accuracy.
Wu, G.D. & Lin, C.T. 2001, 'Noisy speech segmentation with multiband analysis and recurrent neural fuzzy network', Annual Conference of the North American Fuzzy Information Processing Society - NAFIPS, pp. 540-544.
This paper addresses the problem of automatic word boundary detection in the presence of variable-level background noise. Commonly used algorithms for word boundary detection always assume that the background noise level is fixed. In fact, the background noise level may vary during the procedure of recording. In order to solve this problem, we propose the RTF-MiFre-based RSONFIN (a recurrent neural fuzzy network) algorithm. Since the RTF and MiFre parameters can extract useful frequency energy and RSONFIN can process the temporal relations, this RTF-MiFre-based RSONFIN algorithm can find the variation of the background noise level and detect correct word boundaries in the presence of variable background noise level. Our experiment results have shown that the RTF-MiFre-based RSONFIN algorithm has good performance in the presence of variable background noise level presence.
Chung, I.F. & Lin, C.T. 2001, 'A neuro-fuzzy combiner for multiobjective control', Annual Conference of the North American Fuzzy Information Processing Society - NAFIPS, pp. 1384-1389.
This paper proposes a neuro-fuzzy combiner (NFC) with supervised learning capability for solving multiobjective control problems. The proposed NFC can combine n existing low-level controllers in a hierarchical way to form a multiobjective fuzzy controller. It is assumed that each low-level (fuzzy or nonfuzzy) controller has been well designed to serve a particular objective. The role of the NFC is to fuse the n actions decided by the n low-level controllers and determine a proper action acting on the environment (plant) at each time step. Hence, the NFC can combine low-level controller and achieve multiple objectives (goals) at once. Here a NFC can be designed by proposed architecture and supervised learning scheme. Computer simulations have been conducted to illustrate the performance and applicability of the proposed architecture and learning scheme.
Lin, C.T., Chung, I.F. & Lin, J.Y. 2001, 'Multipurpose virtual-reality-based motion simulator', Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, pp. 2846-2851.
Public security has become an important issue everywhere. Especially, the safe manipulation and control of various machines and vehicles has gained special attention such that the authorities keep emphasizing the strict training and censoring of human operators. Currently, such training and censoring process usually relies on the actual machines, equipment, or vehicles in the real sites. This not only has high demands in space, time and cost, but also causes another phase of public security problem. In this connection, the world-wide trend is to tackle the above dilemma by using virtual reality (VR). However, the current researches or products on VR are more matured in the software display part of VR. How to combine 3D VR display with motion platform to achieve the aforementioned training and censoring purposes is an important research issue. This paper focuses on this research issue, and the goal is to develop a multipurpose virtual-reality-based motion simulation system to meet the requirements of public security in training and censoring of human operators.
Duh, F.B. & Lin, C.T. 2001, 'Tracking a maneuvering target using neural fuzzy network', IEEE International Conference on Fuzzy Systems, pp. 1255-1258.
A fast target maneuver detecting and highly accurate tracking technique using a neural fuzzy network based on Kalman filter is proposed in this paper. In the automatic target tracking system, there exists an important and difficult problem: how to detect the maneuvers and fast response to avoid miss-tracking? To solve this problem, neural network and fuzzy algorithm have been issued recently. However, the normal neural networks such as backpropagation networks usually produce the extra problems of low convergence speed and/or large network size, and the fuzzy algorithms are not easy to partition the parameters. To overcome these defects and to make use of neural learning ability, a developed standard Kalman filter with a self-constructing neural fuzzy inference network (KF-SONFIN) algorithm for target tracking is presented in this paper. Without having to change the structure of Kalman filter nor modeling the maneuvering target, SONFIN algorithm, can always find itself an economic network size with a fast learning process. Simulation results show that the KF-SONFIN is superior to the traditional IE and VDF methods in estimation accuracy.
Liu, S.H., Lin, C.T. & Wang, J.J. 2001, 'A model-based fuzzy logic controller for tracking mean arterial pressure', IEEE International Conference on Fuzzy Systems, pp. 1495-1497.
This paper proposes a new noninvasive measurement method for tracking the tendency of mean arterial pressure (MAP) in the radial artery. The designed system consists of a tonometer, a micro syringe device, and a model-based fuzzy logic controller. The proposed control system consists of a linear predictor, and a synthetic fuzzy logic controller (SFLC). The design of the fuzzy rules in each subcontroller is based on the oscillometric principle saying that the arterial vessel has the maximum compliance when the detected vessel volume pulse reaches its maximum amplitude.
Lin, C.T., Chung, I.F., Liann, S.A. & Su, F.Y. 2000, 'Multipurpose virtual-reality-based motion simulator', Proceedings of the World Congress on Intelligent Control and Automation (WCICA), pp. 2699-2704.
Public security has become an important issue everywhere. Especially, the safe manipulation and control of various machines and vehicles has gained special attention such that the authorities keep emphasizing the strict training and censoring of human operators. Currently, such training and censoring process usually relies on the actual machines, equipment, or vehicles in the real sites. This not only has high demands in space, time and cost, but also causes another phase of public security problem. In this connection, the world-wide trend is to tackle the above dilemma by using virtual reality (VR). However, the current researches or products on VR are more matured in the software display part of VR. How to combine 3D VR display with motion platform to achieve the aforementioned training and censoring purposes is an important research issue. This paper focuses on this research issue, and the goal is to develop a multipurpose virtual-reality-based motion simulation system to meet the requirements of public security in training and censoring of human operators.
Wu, S.J. & Lin, C.T. 2000, 'Continuous time-invariant optimal fuzzy tracker design based on local concept approach', IEEE International Conference on Fuzzy Systems, pp. 285-290.
In this paper, we propose a global optimal fuzzy tracking controller for continuous time-invariant fuzzy tracking system to trace the moving or model-following target under infinite-horizon (time) quadratic performance index. We also demonstrate that the stability of the closed-loop fuzzy tracking systems can be ensured by the designed optimal fuzzy tracking controllers. One example is given to illustrate the proposed optimal fuzzy tracker design scheme and to show the proved stability property.
Chung, I.F., Chang, H.H. & Lin, C.T. 1999, 'Fuzzy control of a six-degree motion platform with stability analysis', Proceedings of the IEEE International Conference on Systems, Man and Cybernetics.
This paper addresses the control problem of the six-degree-of-freedom parallel manipulator known as the Stewart platform. This manipulating structure has been well-known for its use as a flight simulator. Recently, it has become an important element in virtual reality (VR) applications for providing motion sensation to users. In this research, we first analyze the hydraulic driving system and the kinematics of the Stewart platform. We then develop a fuzzy controller for the Stewart platform. The stability of the proposed fuzzy control systems are analyzed based on the Popov criterion stability theorem. Experimental results have shown that the designed fuzzy controller can drive the six-degree motion platform accurately, smoothly, and stably.
Wang, Y.J. & Lin, C.T. 1998, 'Runge Kutta Neural Network for identification of continuous systems', Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, pp. 3277-3282.
This paper proposes Runge Kutta Neural Networks (RKNNs) for identification of continuous-time nonlinear systems. These networks are constructed according to the Runge Kutta approximation method. The RKNNs can thus precisely model continuous-time systems and do long-term prediction of system state trajectories. Since the RKNNs model continuous-time systems, they can incorporate available continuous relationship (physical laws) of the identified systems into their structures directly. Also, they are insensitive to the size of sampling interval in prediction. We also show theoretically the superior generalization and long-term prediction capability of the RKNNs over the normal neural networks. A class of novel recursive least square (RLS) algorithms, called nonlinear recursive least square (NRLS) learning algorithms, are developed for the RKNNs. Computer simulations demonstrate the proved properties of the RKNNs.
Juang, C.F. & Lin, C.T. 1998, 'Genetic reinforcement learning through symbiotic evolution for fuzzy controller design', IEEE International Conference on Fuzzy Systems, pp. 1281-1285.
An efficient genetic reinforcement learning algorithm for designing fuzzy controllers is proposed in this paper. The genetic algorithm (GA) adopted in this paper is based upon symbiotic evolution which, when applied to fuzzy controller design, matches well with the local mapping property of a fuzzy rule. Using this Symbiotic-Evolution-based Fuzzy Controller (SEFC) design method, the number of control trials as well as consumed CPU time are reduced considerably as compared to traditional GA-based fuzzy controller design methods and other types of genetic reinforcement learning schemes. The proposed SEFC design method has been applied to the cart-pole balancing system. Efficiency and superiority of the proposed SEFC have been verified from this problem and from the comparisons with traditional GA-based fuzzy systems.
Juang, C.F. & Lin, C.T. 1997, 'Recurrent self-organizing neural fuzzy inference network', IEEE International Conference on Fuzzy Systems, pp. 1369-1374.
A Recurrent Self-Organizing Neural Fuzzy Inference Network (RSONFIN) is proposed in this paper. The RSONFIN is constructed from a series of dynamic fuzzy rules. The temporal relations embedded in the network are built by adding some feedback connections representing the memory elements to a feedforward neural fuzzy network. Each weight as well as node in the RSONFIN has its own meaning and represents a special element in a fuzzy rule. There are no hidden nodes (i.e., no membership functions and fuzzy rules) initially in the RSONFIN. They are created on-line via concurrent structure identification (the construction of dynamic fuzzy if-then rules) and parameter identification (the tuning of the free parameters of membership functions). The structure learning together with the parameter learning forms a fast learning algorithm for building a small, yet powerful, dynamic neural fuzzy network. Simulations on temporal problems are done finally.
Lin, C.T. & Juang, C.F. 1996, 'Adaptive neural fuzzy filter and its applications', IEEE International Conference on Fuzzy Systems, pp. 564-569.
A new kind of nonlinear adaptive filter, the adaptive neural fuzzy filter (ANFF), based upon neural network's learning ability and fuzzy if-then rule structure is proposed in this paper. The ANFF is inherently a feedforward multilayered connectionist network which can learn by itself according to numerical training data or expert knowledge represented by fuzzy if-then rules. The adaptation here includes the construction of fuzzy if-then rules (structure learning), and the tuning of the free parameters of membership functions (parameter learning). There are no hidden nodes (i.e., no membership functions and fuzzy rules) initially, and both the structure learning and parameter learning are performed concurrently as the adaptation proceeds. Two major advantages of the ANFF are: 1) a priori knowledge can be incorporated into the ANFF which makes the fusion of numerical data and linguistic information in the filter possible; and 2) no predetermination, like the number of hidden nodes, must be given, since the ANFF can find its optimal structure and parameters automatically. To demonstrate the performance of the ANFF, two applications, the nonlinear channel equalization and the adaptive noise cancellation, are simulated.
Lin, C.T. & Li, C.P. 1996, 'Temperature control with a neural fuzzy inference network', Proceedings of the Asian Fuzzy Systems Symposium, pp. 91-96.
In this paper, we propose a neural fuzzy inference network (NFIN) suitable for adaptive temperature control of a water bath system. The rules in the NFIN are created and adapted as on-line learning proceeds via simultaneous structure and parameter identification. The NFIN has been applied to a practical water bath temperature control system. The performance of the NFIN is compared to that of the PID controller and fuzzy logic controller (FLC) on the water bath temperature control system. The three control schemes are compared through experimental studies. It's found that the proposed NFIN control scheme has best control performance among the three control schemes.
Lin, C.T., Lin, C.J. & Chung, I.F. 1995, 'Neural fuzzy control of unstable nonlinear systems', Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, pp. 3666-3671.
A Fuzzy Adaptive Learning COntrol Network (FALCON) is proposed for the realization of a fuzzy logic control system. An on-line structure/parameter learning algorithm, called FALCON-ART, can on-line partition the input/output spaces, tune membership functions and find proper fuzzy logic rules dynamically without any a priori knowledge or even any initial information on these. The FALCON-ART requires exact supervised training data for learning. In some real-time applications, exact training data may be expensive or even impossible to obtain. To solve this problem, a Reinforcement Fuzzy Adaptive Learning COntrol Network (RFALCON) is further proposed. By combining a proposed on-line supervised structure/parameter learning technique, the temporal difference method, and the stochastic exploratory algorithm, a on-line supervised structure/parameter learning algorithm, called RFALCON-ART, is proposed for constructing the RFALCON dynamically.
Lin, C.J. & Lin, C.T. 1995, 'Reinforcement learning for ART-based fuzzy adaptive learning control networks', IEEE International Conference on Fuzzy Systems, pp. 1299-1306.
This paper proposes a Reinforcement Fuzzy Adaptive Learning COntrol Network (RFALCON) for solving various reinforcement learning problems. The proposed RFALCON is constructed by integrating two Fuzzy Adaptive Learning COntrol networks (FALCON), each of which is a connectionist model with a feedforward multilayered network developed for the realization of a fuzzy logic controller. An on-line structure/parameter learning algorithm, called RFALCON-ART, is proposed for constructing the RFALCON dynamically. The proposed RFALCON also preserves the advantages of the original FALCON, such as the ability to do on-line partition the input/output spaces, tune membership functions, and find proper fuzzy logic rules. In its initial form, there is no membership function, fuzzy partition, and fuzzy logic rule. They are created and begin to grow as the first reinforcement signal arrives. The users thus need not give it any a priori knowledge or even any initial information on these.
Lin, C.J. & Lin, C.T. 1994, 'ART-based fuzzy adaptive learning control network', pp. 357-362.
This paper addresses the structure and associated on-line learning algorithms of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed Fuzzy Adaptive Learning Control Network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. An on-line structure/parameter learning algorithm, called FALCON-ART, is proposed for constructing the FALCON dynamically. The FALCON-ART can partition the input/output space in a flexible way based on the distribution of the training data. Hence it can avoid the problem of combinatorial growing of partitioned grids in some complex systems. It combines the backpropagation learning scheme for parameter learning and the fuzzy ART algorithm for structure learning. More notably, the FALCON-ART can on-line partition the input/output spaces, tune membership functions and find proper fuzzy logic rules dynamically without any a priori knowledge or even any initial information on these.
Lin, C.J. & Lin, C.T. 1994, 'ART-based fuzzy adaptive learning control network', IEEE International Conference on Fuzzy Systems, pp. 1-5.
This paper addresses the structure and associated on-line learning algorithms of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed Fuzzy Adaptive Learning COntrol Network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. An on-line structure/parameter learning algorithm, called FALCON-ART, is proposed for constructing the FALCON dynamically. The FALCON-ART can partition the input/output space in a flexible way based on the distribution of the training data. Hence it can avoid the problem of combinatorial growing of partitioned grids in some complex systems. It combines the backpropagation learning scheme for parameter learning and the fuzzy ART algorithm for structure learning. More notably, the FALCON-ART can on-line partition the input/output spaces, tune membership functions and find proper fuzzy logic rules dynamically without any a priori knowledge or even any initial information on these.
Lin, C.T. 1994, 'FALCON: A fuzzy adaptive learning control network', pp. 228-232.
This paper proposes a Reinforcement Fuzzy Adaptive Learning COntrol Network (RFALCON) for solving various reinforcement learning problems. The proposed RFALCON is constructed by integrating two Fuzzy Adaptive Learning COntrol networks (FALCON), each of which is a connectionist model with a feedforward multilayered network developed for the realization of a fuzzy logic controller. An on-line structure/parameter learning algorithm, called RFALCON-ART, is proposed for constructing the RFALCON dynamically. The proposed R-FALCON also preserves the advantages of the original FALCON, such as the ability to do on-line partition the input/output spaces, tune membership functions, and find proper fuzzy logic rules. In its initial form, there is no membership function, fuzzy partition, and fuzzy logic rule. They are created and begin to grow as the first reinforcement signal arrives. The users thus need not give it any a priori knowledge or even any initial information on these.
Lin, C.T. & Lee, C.S.G. 1993, 'Reinforcement structure/parameter learning for neural-network-based fuzzy logic control systems', 1993 IEEE International Conference on Fuzzy Systems, pp. 88-93.
This paper proposes a Reinforcement Neural-Network-Based Fuzzy Logic Control System (RNN-FLCS) for solving various reinforcement learning problems. The proposed RNN-FLCS is best applied to learning environments where obtaining exact training data is expensive. It is constructed by integrating two Neural-Network-Based Fuzzy Logic Controllers (NN-FLCs), each of which is a connectionist model with a feedforward multi-layered network developed for the realization of a fuzzy logic controller. One NN-FLC functions as a fuzzy predictor and the other as a fuzzy controller. Using the temporal difference prediction method, the fuzzy predictor can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the fuzzy controller. The fuzzy controller performs a stochastic exploratory algorithm to adapt itself according to the internal reinforcement signal. During the learning process, the proposed RNN-FLCS can construct a fuzzy logic control system automatically and dynamically through a reward-penalty signal or through very simple fuzzy information feedback; both structure learning and parameter learning are performed simultaneously in the two NN-FLCs using the fuzzy similarity measure. Simulation results are presented to illustrate the performance and applicability of the proposed RNN-FLCS.
Lin, C.T. & Lee, C.S.G. 1992, 'Real-time supervised structure/parameter learning for fuzzy neural network', pp. 1283-1291.
The authors propose a real-time supervised structure and parameter learning algorithm for constructing fuzzy neural networks (FNNs) automatically and dynamically. This algorithm combines the backpropagation learning scheme for the parameter learning and a novel fuzzy similarity measure for the structure learning. The fuzzy similarity measure is a new tool to determine the degree to which two fuzzy sets are equal. The FNN is a feedforward multi-layered network which integrates the basic elements and functions of a traditional fuzzy logic controller into a connectionist structure which has distributed learning abilities. The structure learning decides the proper connection types and the number of hidden units which represent fuzzy logic rules and the number of fuzzy partitions. The parameter learning adjusts the node and link parameters which represent the membership functions. The proposed supervised learning algorithm provides an efficient way for constructing a FNN in real time. Simulation results are presented to illustrate the performance and applicability of the proposed learning algorithm.

Journal articles

Liang, C., Lin, C.T., Yao, S.N., Chang, W.S., Liu, Y.C. & Chen, S.A. 2017, 'Visual attention and association: An electroencephalography study in expert designers', Design Studies, vol. 48, pp. 76-95.
View/Download from: Publisher's site
© 2016 Elsevier LtdExtant research on the visual attention and association of designers is limited, and scientific evidence differentiating among the effects of diverse visual stimuli on design thinking is insufficient. The current study invited 12 healthy expert designers and analysed their experiences of visual attention and association in addition to exploring the differences caused by three types of pictorial representation. The results of this electroencephalography (EEG) experiment indicated that the frontoparietal region was particularly activated when the designers engaged in visual attention tasks, whereas the brainwaves were particularly activated in the distributed prefrontal, frontocentral, and parietooccipital regions during the visual association tasks. In addition, there were no significant differences in the brainwave energy resulting from the three types of pictorial representation applied in this study. The research outcomes linking design studies to cognitive neuroscience establish a concrete foundation for developing future applied research and diverse educational practices.
Yao, S.N., Lin, C.T., King, J.T., Liu, Y.C. & Liang, C. 2017, 'Learning in the visual association of novice and expert designers', Cognitive Systems Research, vol. 43, pp. 76-88.
View/Download from: Publisher's site
© 2017 Elsevier B.V.Designers are adept at determining similarities between previously seen objects and new creations using visual association. However, extant research on the visual association of designers and the differences between expert and novice designers when they engage in the visual association task are scant. Using electroencephalography (EEG), this study attempted to narrow this research gap. Sixteen healthy designers—eight experts and eight novices—were recruited, and asked to perform visual association while EEG signals were acquired, subsequently analysed using independent component analysis. The results indicated that strong connectivity was observed among the prefrontal, frontal, and cingulate cortices, and the default mode network. The experts used both hemispheres and executive functions to support their association tasks, whereas the novices mainly used their right hemisphere and memory retrieval functions. The visual association of experts appeared to be more goal-directed than that of the novices. Accordingly, designing and implementing authentic and goal-directed activities for improving the executive functions of the prefrontal cortex and default mode network are critical for design educators and creativity researchers.
Liu, Y.-.T., Pal, N.R., Marathe, A. & Lin, C.-.T. 2017, 'Weighted Fuzzy Dempster-Shafer Framework for Multi-Modal Information Integration', IEEE Transactions on Fuzzy Systems, pp. 1-1.
View/Download from: Publisher's site
Chuang, S.W., Chuang, C.H., Yu, Y.H., King, J.T. & Lin, C.T. 2016, 'EEG Alpha and Gamma Modulators Mediate Motion Sickness-Related Spectral Responses.', International journal of neural systems, vol. 26, no. 2, p. 1650007.
View/Download from: UTS OPUS
Motion sickness (MS) is a common experience of travelers. To provide insights into brain dynamics associated with MS, this study recruited 19 subjects to participate in an electroencephalogram (EEG) experiment in a virtual-reality driving environment. When riding on consecutive winding roads, subjects experienced postural instability and sensory conflict between visual and vestibular stimuli. Meanwhile, subjects rated their level of MS on a six-point scale. Independent component analysis (ICA) was used to separate the filtered EEG signals into maximally temporally independent components (ICs). Then, reduced logarithmic spectra of ICs of interest, using principal component analysis, were decomposed by ICA again to find spectrally fixed and temporally independent modulators (IMs). Results demonstrated that a higher degree of MS accompanied increased activation of alpha (r = 0.421) and gamma (r =0.478) IMs across remote-independent brain processes, covering motor, parietal and occipital areas. This co-modulatory spectral change in alpha and gamma bands revealed the neurophysiological demand to regulate conflicts among multi-modal sensory systems during MS.
Yu, Y.H., Lu, S.W., Chuang, C.H., King, J.T., Chang, C.L., Chen, S.A., Chen, S.F. & Lin, C.T. 2016, 'An inflatable and wearable wireless system for making 32-channel electroencephalogram measurements', IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 24, no. 7, pp. 806-813.
View/Download from: UTS OPUS or Publisher's site
© 2001-2011 IEEE.Potable electroencephalography (EEG) devices have become critical for important research. They have various applications, such as in brain-computer interfaces (BCI). Numerous recent investigations have focused on the development of dry sensors, but few concern the simultaneous attachment of high-density dry sensors to different regions of the scalp to receive qualified EEG signals from hairy sites. An inflatable and wearable wireless 32-channel EEG device was designed, prototyped, and experimentally validated for making EEG signal measurements; it incorporates spring-loaded dry sensors and a novel gasbag design to solve the problem of interference by hair. The cap is ventilated and incorporates a circuit board and battery with a high-tolerance wireless (Bluetooth) protocol and low power consumption characteristics. The proposed system provides a 500/250 Hz sampling rate, and 24 bit EEG data to meet the BCI system data requirement. Experimental results prove that the proposed EEG system is effective in measuring audio event-related potential, measuring visual event-related potential, and rapid serial visual presentation. Results of this work demonstrate that the proposed EEG cap system performs well in making EEG measurements and is feasible for practical applications.
Lin, C.-.T., Chuang, C.-.H., Kerick, S., Mullen, T., Jung, T.-.P., Ko, L.-.W., Chen, S.-.A., King, J.-.T. & McDowell, K. 2016, 'Mind-Wandering Tends to Occur under Low Perceptual Demands during Driving.', Sci Rep, vol. 6, p. 21353.
View/Download from: UTS OPUS or Publisher's site
Fluctuations in attention behind the wheel poses a significant risk for driver safety. During transient periods of inattention, drivers may shift their attention towards internally-directed thoughts or feelings at the expense of staying focused on the road. This study examined whether increasing task difficulty by manipulating involved sensory modalities as the driver detected the lane-departure in a simulated driving task would promote a shift of brain activity between different modes of processing, reflected by brain network dynamics on electroencephalographic sources. Results showed that depriving the driver of salient sensory information imposes a relatively more perceptually-demanding task, leading to a stronger activation in the task-positive network. When the vehicle motion feedback is available, the drivers may rely on vehicle motion to perceive the perturbations, which frees attentional capacity and tends to activate the default mode network. Such brain network dynamics could have major implications for understanding fluctuations in driver attention and designing advance driver assistance systems.
Liu, Y.T., Lin, Y.Y., Wu, S.L., Chuang, C.H. & Lin, C.T. 2016, 'Brain Dynamics in Predicting Driving Fatigue Using a Recurrent Self-Evolving Fuzzy Neural Network', IEEE Transactions on Neural Networks and Learning Systems, vol. 27, no. 2, pp. 347-360.
View/Download from: UTS OPUS or Publisher's site
© 2012 IEEE. This paper proposes a generalized prediction system called a recurrent self-evolving fuzzy neural network (RSEFNN) that employs an on-line gradient descent learning rule to address the electroencephalography (EEG) regression problem in brain dynamics for driving fatigue. The cognitive states of drivers significantly affect driving safety; in particular, fatigue driving, or drowsy driving, endangers both the individual and the public. For this reason, the development of brain-computer interfaces (BCIs) that can identify drowsy driving states is a crucial and urgent topic of study. Many EEG-based BCIs have been developed as artificial auxiliary systems for use in various practical applications because of the benefits of measuring EEG signals. In the literature, the efficacy of EEG-based BCIs in recognition tasks has been limited by low resolutions. The system proposed in this paper represents the first attempt to use the recurrent fuzzy neural network (RFNN) architecture to increase adaptability in realistic EEG applications to overcome this bottleneck. This paper further analyzes brain dynamics in a simulated car driving task in a virtual-reality environment. The proposed RSEFNN model is evaluated using the generalized cross-subject approach, and the results indicate that the RSEFNN is superior to competing models regardless of the use of recurrent or nonrecurrent structures.
Huang, K.-.C., Huang, T.-.Y., Chuang, C.-.H., King, J.-.T., Wang, Y.-.K., Lin, C.-.T. & Jung, T.-.P. 2016, 'An EEG-Based Fatigue Detection and Mitigation System.', International journal of neural systems, vol. 26, no. 4, p. 1650018.
View/Download from: UTS OPUS or Publisher's site
Research has indicated that fatigue is a critical factor in cognitive lapses because it negatively affects an individual's internal state, which is then manifested physiologically. This study explores neurophysiological changes, measured by electroencephalogram (EEG), due to fatigue. This study further demonstrates the feasibility of an online closed-loop EEG-based fatigue detection and mitigation system that detects physiological change and can thereby prevent fatigue-related cognitive lapses. More importantly, this work compares the efficacy of fatigue detection and mitigation between the EEG-based and a nonEEG-based random method. Twelve healthy subjects participated in a sustained-attention driving experiment. Each participant's EEG signal was monitored continuously and a warning was delivered in real-time to participants once the EEG signature of fatigue was detected. Study results indicate suppression of the alpha- and theta-power of an occipital component and improved behavioral performance following a warning signal; these findings are in line with those in previous studies. However, study results also showed reduced warning efficacy (i.e. increased response times (RTs) to lane deviations) accompanied by increased alpha-power due to the fluctuation of warnings over time. Furthermore, a comparison of EEG-based and nonEEG-based random approaches clearly demonstrated the necessity of adaptive fatigue-mitigation systems, based on a subject's cognitive level, to deliver warnings. Analytical results clearly demonstrate and validate the efficacy of this online closed-loop EEG-based fatigue detection and mitigation mechanism to identify cognitive lapses that may lead to catastrophic incidents in countless operational environments.
Ding, W.P., Lin, C.T., Prasad, M., Chen, S.B. & Guan, Z.J. 2016, 'Attribute Equilibrium Dominance Reduction Accelerator (DCCAEDR) Based on Distributed Coevolutionary Cloud and Its Application in Medical Records', IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 46, no. 3, pp. 384-400.
View/Download from: Publisher's site
© 2013 IEEE. Aimed at the tremendous challenge of attribute reduction for big data mining and knowledge discovery, we propose a new attribute equilibrium dominance reduction accelerator (DCCAEDR) based on the distributed coevolutionary cloud model. First, the framework of N-populations distributed coevolutionary MapReduce model is designed to divide the entire population into N subpopulations, sharing the reward of different subpopulations' solutions under a MapReduce cloud mechanism. Because the adaptive balancing between exploration and exploitation can be achieved in a better way, the reduction performance is guaranteed to be the same as those using the whole independent data set. Second, a novel Nash equilibrium dominance strategy of elitists under the N bounded rationality regions is adopted to assist the subpopulations necessary to attain the stable status of Nash equilibrium dominance. This further enhances the accelerator's robustness against complex noise on big data. Third, the approximation parallelism mechanism based on MapReduce is constructed to implement rule reduction by accelerating the computation of attribute equivalence classes. Consequently, the entire attribute reduction set with the equilibrium dominance solution can be achieved. Extensive simulation results have been used to illustrate the effectiveness and robustness of the proposed DCCAEDR accelerator for attribute reduction on big data. Furthermore, the DCCAEDR is applied to solve attribute reduction for traditional Chinese medical records and to segment cortical surfaces of the neonatal brain 3-D-MRI records, and the DCCAEDR shows the superior competitive results, when compared with the representative algorithms.
Li, D.L., Prasad, M., Lin, C.T. & Chang, J.Y. 2016, 'Self-adjusting feature maps network and its applications', Neurocomputing, vol. 207, pp. 78-94.
View/Download from: Publisher's site
This paper, proposes a novel artificial neural network, called self-adjusting feature map (SAM), and develop its unsupervised learning ability with self-adjusting mechanism. The trained network structure of representative connected neurons not only displays the spatial relation of the input data distribution but also quantizes the data well. The SAM can automatically isolate a set of connected neurons, in which, the used number of the sets may indicate the number of clusters. The idea of self-adjusting mechanism is based on combining of mathematical statistics and neurological advantages and retreat of waste. In the training process, for each representative neuron has are three phases, growth, adaptation and decline. The network of representative neurons, first create the necessary neurons according to the local density of the input data in the growth phase. In the adaption phase, it adjusts neighborhood neuron pair's connected/disconnected topology constantly according to the statistics of input feature data. Finally, the unnecessary neurons of the network are merged or remove in the decline phase. In this paper, we exploit the SAM to handle some peculiar cases that cannot be handled easily by classical unsupervised learning networks such as self-organizing map (SOM) network. The remarkable characteristics of the SAM can be seen on various real world cases in the experimental results.
Gupta, P., Lin, C.T., Mehlawat, M.K. & Grover, N. 2016, 'A New Method for Intuitionistic Fuzzy Multiattribute Decision Making', IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 46, no. 9, pp. 1167-1179.
View/Download from: Publisher's site
© 2013 IEEE.In this paper, we study the multiattribute decision-making (MADM) problem with intuitionistic fuzzy values that represent information regarding alternatives on the attributes. Assuming that the weight information of the attributes is not known completely, we use an approach that utilizes the relative comparisons based on the advantage and disadvantage scores of the alternatives obtained on each attribute. The relative comparison of the intuitionistic fuzzy values in this research use all the three parameters, namely membership degree ('the more the better'), nonmembership degree ('the less the better'), and hesitancy degree ('the less the better'), thereby leading to the tradeoff values of all the three parameters. The score functions (advantage and disadvantage scores) used for this purpose are based on the positive contributions of these parameters, wherever applicable. Furthermore, these scores are used to obtain the strength and weakness scores leading to the satisfaction degrees of the alternatives. The optimal weights of the attributes are determined using a multiobjective optimization model that simultaneously maximizes the satisfaction degree of each alternative. The optimal solution is used for ranking and selecting the best alternative on the basis of the overall attribute values. To validate the proposed methodology, we present a numerical illustration of a real-world case. The methodology is further extended to treat MADM problem with interval-valued intuitionistic fuzzy information. Finally, a thorough comparison is done to demonstrate the advantages of the solution methodology over the existing methods used for the intuitionistic fuzzy MADM problems.
Wu, S.-.L., Liu, Y.-.T., Hsieh, T.-.Y., Lin, Y.-.Y., Chen, C.-.Y., Chuang, C.-.H. & Lin, C.-.T. 2016, 'Fuzzy Integral with Particle Swarm Optimization for a Motor-Imagery-based Brain-Computer Interface', IEEE Transactions on Fuzzy Systems, pp. 1-1.
View/Download from: Publisher's site
Yu, Y.-.H., Chen, S.-.H., Chang, C.-.L., Lin, C.-.T., Hairston, W.D. & Mrozek, R.A. 2016, 'New Flexible Silicone-Based EEG Dry Sensor Material Compositions Exhibiting Improvements in Lifespan, Conductivity, and Reliability.', Sensors (Basel), vol. 16, no. 11.
View/Download from: Publisher's site
This study investigates alternative material compositions for flexible silicone-based dry electroencephalography (EEG) electrodes to improve the performance lifespan while maintaining high-fidelity transmission of EEG signals. Electrode materials were fabricated with varying concentrations of silver-coated silica and silver flakes to evaluate their electrical, mechanical, and EEG transmission performance. Scanning electron microscope (SEM) analysis of the initial electrode development identified some weak points in the sensors' construction, including particle pull-out and ablation of the silver coating on the silica filler. The newly-developed sensor materials achieved significant improvement in EEG measurements while maintaining the advantages of previous silicone-based electrodes, including flexibility and non-toxicity. The experimental results indicated that the proposed electrodes maintained suitable performance even after exposure to temperature fluctuations, 85% relative humidity, and enhanced corrosion conditions demonstrating improvements in the environmental stability. Fabricated flat (forehead) and acicular (hairy sites) electrodes composed of the optimum identified formulation exhibited low impedance and reliable EEG measurement; some initial human experiments demonstrate the feasibility of using these silicone-based electrodes for typical lab data collection applications.
Cao, Z., Lin, C.-.T., Chuang, C.-.H., Lai, K.-.L., Yang, A.C., Fuh, J.-.L. & Wang, S.-.J. 2016, 'Resting-state EEG power and coherence vary between migraine phases.', J Headache Pain, vol. 17, no. 1, p. 102.
View/Download from: Publisher's site
BACKGROUND: Migraine is characterized by a series of phases (inter-ictal, pre-ictal, ictal, and post-ictal). It is of great interest whether resting-state electroencephalography (EEG) is differentiable between these phases. METHODS: We compared resting-state EEG energy intensity and effective connectivity in different migraine phases using EEG power and coherence analyses in patients with migraine without aura as compared with healthy controls (HCs). EEG power and isolated effective coherence of delta (1-3.5&nbsp;Hz), theta (4-7.5&nbsp;Hz), alpha (8-12.5&nbsp;Hz), and beta (13-30&nbsp;Hz) bands were calculated in the frontal, central, temporal, parietal, and occipital regions. RESULTS: Fifty patients with episodic migraine (1-5 headache days/month) and 20 HCs completed the study. Patients were classified into inter-ictal, pre-ictal, ictal, and post-ictal phases (n=22, 12, 8, 8, respectively), using 36-h criteria. Compared to HCs, inter-ictal and ictal patients, but not pre- or post-ictal patients, had lower EEG power and coherence, except for a higher effective connectivity in fronto-occipital network in inter-ictal patients (p<.05). Compared to data obtained from the inter-ictal group, EEG power and coherence were increased in the pre-ictal group, with the exception of a lower effective connectivity in fronto-occipital network (p<.05). Inter-ictal and ictal patients had decreased EEG power and coherence relative to HCs, which were "normalized" in the pre-ictal or post-ictal groups. CONCLUSION: Resting-state EEG power density and effective connectivity differ between migraine phases and provide an insight into the complex neurophysiology of migraine.
Lin, C.T. & Garibaldi, J.J. 2016, 'Editorial', IEEE Transactions on Fuzzy Systems, vol. 24, no. 6, pp. 1257-1258.
View/Download from: Publisher's site
Liu, Y.T., Chuang, C.H., Wang JM & Lin, C.T. 2016, 'Changes in Alertness and the EEG Effective Connectivity in a Sustained-Attention Driving Task'.
Singh, J., Prasad, M., Prasad, O.K., Meng Joo, E., Saxena, A.K. & Lin, C.-.T. 2016, 'A Novel Fuzzy Logic Model for Pseudo-Relevance Feedback-Based Query Expansion', International Journal of Fuzzy Systems, vol. 18, no. 6, pp. 980-989.
View/Download from: Publisher's site
Liu, Y.T., Lin, C., Chuang, C.H., Wang, Y.K., Huang, S.H., King, J.T., Chen, S.A. & Lu, S.W. 2016, 'Novel Neurotechnology and Computational Intelligence Method Applied to EEG-based Brain-Computer Interfaces'.
Kaiwartya, O., Abdullah, A.H., Cao, Y., Altameem, A., Prasad, M., Lin, C.-.T. & Liu, X. 2016, 'Internet of Vehicles: Motivation, Layered Architecture, Network Model, Challenges, and Future Aspects', IEEE Access, vol. 4, pp. 5356-5373.
View/Download from: Publisher's site
Wu, D., Lawhern, V.J., Gordon, S., Lance, B.J. & Lin, C.-.T. 2016, 'Driver Drowsiness Estimation from EEG Signals Using Online Weighted Adaptation Regularization for Regression (OwARR)', IEEE Transactions on Fuzzy Systems, pp. 1-1.
View/Download from: Publisher's site
Lin, C., Hajinoroozi, M., Mao, Z., Jung, T.P. & Huang, Y.F. 2016, 'EEG- based prediction of drivers cognitive performance by deep convolutional neural network'.
Dzeng, R.-.J., Lin, C.-.T. & Fang, Y.-.C. 2016, 'Using eye-tracker to compare search patterns between experienced and novice workers for site hazard identification', Safety Science, vol. 82, pp. 56-67.
View/Download from: Publisher's site
Lin, C.-.T., Chiu, T.-.C. & Gramann, K. 2015, 'EEG correlates of spatial orientation in the human retrosplenial complex.', Neuroimage, vol. 120, pp. 123-132.
View/Download from: UTS OPUS or Publisher's site
Studies on spatial navigation reliably demonstrate that the retrosplenial complex (RSC) plays a pivotal role for allocentric spatial information processing by transforming egocentric and allocentric spatial information into the respective other spatial reference frame (SRF). While more and more imaging studies investigate the role of the RSC in spatial tasks, high temporal resolution measures such as electroencephalography (EEG) are missing. To investigate the function of the RSC in spatial navigation with high temporal resolution we used EEG to analyze spectral perturbations during navigation based on allocentric and egocentric SRF. Participants performed a path integration task in a clearly structured virtual environment providing allothetic information. Continuous EEG recordings were decomposed by independent component analysis (ICA) with subsequent source reconstruction of independent time source series using equivalent dipole modeling. Time-frequency transformation was used to investigate reference frame-specific orientation processes during navigation as compared to a control condition with identical visual input but no orientation task. Our results demonstrate that navigation based on an egocentric reference frame recruited a network including the parietal, motor, and occipital cortices with dominant perturbations in the alpha band and theta modulation in frontal cortex. Allocentric navigation was accompanied by performance-related desynchronization of the 8-13 Hz frequency band and synchronization in the 12-14 Hz band in the RSC. The results support the claim that the retrosplenial complex is central to translating egocentric spatial information into allocentric reference frames. Modulations in different frequencies with different time courses in the RSC further provide first evidence of two distinct neural processes reflecting translation of spatial information based on distinct reference frames and the computation of heading changes.
Huang, C.S., Pal, N.R., Chuang, C.H. & Lin, C.T. 2015, 'Identifying changes in EEG information transfer during drowsy driving by transfer entropy.', Frontiers in human neuroscience, vol. 9, p. 570.
View/Download from: UTS OPUS
Drowsy driving is a major cause of automobile accidents. Previous studies used neuroimaging based approaches such as analysis of electroencephalogram (EEG) activities to understand the brain dynamics of different cortical regions during drowsy driving. However, the coupling between brain regions responding to this vigilance change is still unclear. To have a comprehensive understanding of neural mechanisms underlying drowsy driving, in this study we use transfer entropy, a model-free measure of effective connectivity based on information theory. We investigate the pattern of information transfer between brain regions when the vigilance level, which is derived from the driving performance, changes from alertness to drowsiness. Results show that the couplings between pairs of frontal, central, and parietal areas increased at the intermediate level of vigilance, which suggests that an enhancement of the cortico-cortical interaction is necessary to maintain the task performance and prevent behavioral lapses. Additionally, the occipital-related connectivity magnitudes monotonically decreases as the vigilance level declines, which further supports the cortical gating of sensory stimuli during drowsiness. Neurophysiological evidence of mutual relationships between brain regions measured by transfer entropy might enhance the understanding of cortico-cortical communication during drowsy driving.
Wang, Y.K., Jung, T.P. & Lin, C.T. 2015, 'EEG-Based Attention Tracking During Distracted Driving', IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 23, no. 6, pp. 1085-1094.
View/Download from: Publisher's site
&copy; 2001-2011 IEEE. Distracted driving might lead to many catastrophic consequences. Developing a countermeasure to track drivers' focus of attention (FOA) and engagement of operators in dual (multi)-tasking conditions is thus imperative. Ten healthy volunteers participated in a dual-task experiment that comprised two tasks: a lane-keeping driving task and a mathematical problem-solving task (e.g., 24+15=37?) during which their electroencephalogram (EEG) and behaviors were concurrently recorded. Independent component analysis (ICA) was employed as a spatial filter to separate the contributions of independent sources from the recorded EEG data. The power spectra of six components (i.e., frontal, central, parietal, occipital, left motor, and right motor) extracted from single-task conditions were fed into support vector machine (SVM) based on the radial basis function (RBF) kernel to build an FOA assessment system. The system achieved 84.6&plusmn;5.8% and 86.2&plusmn;5.4% classification accuracies in detecting the participants' FOAs on the math versus driving tasks, respectively. This FOA assessment system was then applied to evaluate participants' FOAs during dual-task conditions. The detected FOAs revealed that participants' cognitive attention and strategies dynamically changed between tasks to optimize the overall performance, as attention was limited and competed. The empirical results of this study demonstrate the feasibility of a practical system to continuously estimating cognitive attention through EEG spectra.
Chuang, C.H., Huang, C.S., Ko, L.W. & Lin, C.T. 2015, 'An EEG-based perceptual function integration network for application to drowsy driving', Knowledge-Based Systems, vol. 80, pp. 143-152.
View/Download from: UTS OPUS or Publisher's site
&copy; 2015 Elsevier B.V. All rights reserved. Drowsy driving is among the most critical causes of fatal crashes. Thus, the development of an effective algorithm for detecting a driver's cognitive state demands immediate attention. For decades, studies have observed clear evidence using electroencephalography that the brain's rhythmic activities fluctuate from alertness to drowsiness. Recognition of this physiological signal is the major consideration of neural engineering for designing a feasible countermeasure. This study proposed a perceptual function integration system which used spectral features from multiple independent brain sources for application to recognize the driver's vigilance state. The analysis of brain spectral dynamics demonstrated physiological evidenced that the activities of the multiple cortical sources were highly related to the changes of the vigilance state. The system performances showed a robust and improved accuracy as much as 88% higher than any of results performed by a single-source approach.
Li, S.Y., Chen, S.A., Lin, C.T., Ko, L.W., Yang, C.H. & Chen, H.H. 2015, 'Generalized synchronization of nonlinear chaotic systems through natural bioinspired controlling strategy', Abstract and Applied Analysis, vol. 2015.
View/Download from: Publisher's site
&copy; 2015 Shih-Yu Li et al. A novel bioinspired control strategy design is proposed for generalized synchronization of nonlinear chaotic systems, combining the bioinspired stability theory, fuzzy modeling, and a novel, simple-form Lyapunov control function design of derived high efficient, heuristic and bioinspired controllers. Three main contributions are concluded: (1) apply the bioinspired stability theory to further analyze the stability of fuzzy error systems; the high performance of controllers has been shown in previous study by Li and Ge 2009, (2) a new Lyapunov control function based on bioinspired stability theory is designed to achieve synchronization without using traditional LMI method, which is a simple linear homogeneous function of states and the process of designing controller to synchronize two fuzzy chaotic systems becomes much simpler, and (3) three different situations of synchronization are proposed; classical master and slave Lorenz systems, slave Chen's system, and Rossler's system as functional system are illustrated to further show the effectiveness and feasibility of our novel strategy. The simulation results show that our novel control strategy can be applied to different and complicated control situations with high effectiveness.
Liao, S.H., Hsieh, J.G., Chang, J.Y. & Lin, C.T. 2014, 'Training neural networks via simplified hybrid algorithm mixing Nelder–Mead and particle swarm optimization methods', Soft Computing, vol. 19, no. 3, pp. 679-689.
View/Download from: Publisher's site
&copy; 2014, Springer-Verlag Berlin Heidelberg. In this paper, a new and simplified hybrid algorithm mixing the simplex method of Nelder and Mead (NM) and particle swarm optimization algorithm (PSO), abbreviated as SNM-PSO, is proposed for the training of the parameters of the Artificial Neural Network (ANN). Our method differs from other hybrid PSO methods in that, n+1 particles, where n is the dimension of the search space, are randomly selected (without sorting), at each iteration of the proposed algorithm for use as the initial vertices of the NM algorithm, and each such particle is replaced by the corresponding final vertex after executing the NM algorithm. All the particles are then updated using the standard PSO algorithm. Our proposed method is simpler than other similar hybrid PSO methods and places more emphasis on the exploration of the search space. Some simulation problems will be provided to compare the performances of the proposed method with PSO and other similar hybrid PSO methods in training an ANN. These simulations show that the proposed method outperforms the other compared methods.
Yu, Y.H., Lu, S.W., Liao, L.D. & Lin, C.T. 2014, 'Design, fabrication, and experimental validation of novel flexible silicon-based dry sensors for electroencephalography signal measurements', IEEE Journal of Translational Engineering in Health and Medicine, vol. 2.
View/Download from: Publisher's site
&copy; 2014 IEEE. Many commercially available electroencephalography (EEG) sensors, including conventional wet and dry sensors, can cause skin irritation and user discomfort owing to the foreign material. The EEG products, especially sensors, highly prioritize the comfort level during devices wear. To overcome these drawbacks for EEG sensors, this paper designs Societe Generale de Surveillance S A c(SGS)-certified, silicon-based dry-contact EEG sensors (SBDSs) for EEG signal measurements. According to the SGS testing report, SBDSs extract does not irritate skin or induce noncytotoxic effects on L929 cells according to ISO10993-5. The SBDS is also lightweight, flexible, and nonirritating to the skin, as well as capable of easily fitting to scalps without any skin preparation or use of a conductive gel. For forehead and hairy sites, EEG signals can be measured reliably with the designed SBDSs. In particular, for EEG signal measurements at hairy sites, the acicular and flexible design of SBDS can push the hair aside to achieve satisfactory scalp contact, as well as maintain low skin-electrode interface impedance. Results of this paper demonstrate that the proposed sensors perform well in the EEG measurements and are feasible for practical applications.
Wang, Y.T., Huang, K.C., Wei, C.S., Huang, T.Y., Ko, L.W., Lin, C.T., Cheng, C.K. & Jung, T.P. 2014, 'Developing an EEG-based on-line closed-loop lapse detection and mitigation system', Frontiers in Neuroscience, vol. 8, no. OCT.
View/Download from: Publisher's site
&copy; 2014 Wang, Huang, Wei, Huang, Ko, Lin, Cheng and Jung. In America, 60% of adults reported that they have driven a motor vehicle while feeling drowsy, and at least 15-20% of fatal car accidents are fatigue-related. This study translates previous laboratory-oriented neurophysiological research to design, develop, and test an On-line Closed-loop Lapse Detection and Mitigation (OCLDM) System featuring a mobile wireless dry-sensor EEG headgear and a cell-phone based real-time EEG processing platform. Eleven subjects participated in an event-related lane-keeping task, in which they were instructed to manipulate a randomly deviated, fixed-speed cruising car on a 4-lane highway. This was simulated in a 1st person view with an 8-screen and 8-projector immersive virtual-reality environment. When the subjects experienced lapses or failed to respond to events during the experiment, auditory warning was delivered to rectify the performance decrements. However, the arousing auditory signals were not always effective. The EEG spectra exhibited statistically significant differences between effective and ineffective arousing signals, suggesting that EEG spectra could be used as a countermeasure of the efficacy of arousing signals. In this on-line pilot study, the proposed OCLDM System was able to continuously detect EEG signatures of fatigue, deliver arousing warning to subjects suffering momentary cognitive lapses, and assess the efficacy of the warning in near real-time to rectify cognitive lapses. The on-line testing results of the OCLDM System validated the efficacy of the arousing signals in improving subjects' response times to the subsequent lane-departure events. This study may lead to a practical on-line lapse detection and mitigation system in real-world environments.
Huang, C.S., Lin, C.L., Ko, L.W., Liu, S.Y., Su, T.P. & Lin, C.T. 2014, 'Knowledge-based identification of sleep stages based on two forehead electroencephalogram channels', Frontiers in Neuroscience, vol. 8, no. SEP.
View/Download from: Publisher's site
&copy; 2014 Huang, Lin, Ko, Liu, Su and Lin. Sleep quality is important, especially given the considerable number of sleep-related pathologies. The distribution of sleep stages is a highly effective and objective way of quantifying sleep quality. As a standard multi-channel recording used in the study of sleep, polysomnography (PSG) is a widely used diagnostic scheme in sleep medicine. However, the standard process of sleep clinical test, including PSG recording and manual scoring, is complex, uncomfortable, and time-consuming. This process is difficult to implement when taking the whole PSG measurements at home for general healthcare purposes. This work presents a novel sleep stage classification system, based on features from the two forehead EEG channels FP1 and FP2. By recording EEG from forehead, where there is no hair, the proposed system can monitor physiological changes during sleep in a more practical way than previous systems. Through a headband or self-adhesive technology, the necessary sensors can be applied easily by users at home. Analysis results demonstrate that classification performance of the proposed system overcomes the individual differences between different participants in terms of automatically classifying sleep stages. Additionally, the proposed sleep stage classification system can identify kernel sleep features extracted from forehead EEG, which are closely related with sleep clinician's expert knowledge. Moreover, forehead EEG features are classified into five sleep stages by using the relevance vector machine. In a leave-one-subject-out cross validation analysis, we found our system to correctly classify five sleep stages at an average accuracy of 76.7 &plusmn; 4.0 (SD) % [average kappa 0.68 &plusmn; 0.06 (SD)]. Importantly, the proposed sleep stage classification system using forehead EEG features is a viable alternative for measuring EEG signals at home easily and conveniently to evaluate sleep quality reliably, ultimately improving public healthcare...
Chakraborty, R., Lin, C.-.T. & Pal, N.R. 2014, 'Sensor (group feature) selection with controlled redundancy in a connectionist framework.', International journal of neural systems, vol. 24, no. 6, p. 1450021.
View/Download from: Publisher's site
For many applications, to reduce the processing time and the cost of decision making, we need to reduce the number of sensors, where each sensor produces a set of features. This sensor selection problem is a generalized feature selection problem. Here, we first present a sensor (group-feature) selection scheme based on Multi-Layered Perceptron Networks. This scheme sometimes selects redundant groups of features. So, we propose a selection scheme which can control the level of redundancy between the selected groups. The idea is general and can be used with any learning scheme. We have demonstrated the effectiveness of our scheme on several data sets. In this context, we define different measures of sensor dependency (dependency between groups of features). We have also presented an alternative learning scheme which is more effective than our old scheme. The proposed scheme is also adapted to radial basis function (RBS) network. The advantages of our scheme are threefold. It looks at all the groups together and hence can exploit nonlinear interaction between groups, if any. Our scheme can simultaneously select useful groups as well as learn the underlying system. The level of redundancy among groups can also be controlled.
Khoo, I.H., Reddy, H.C., Van, L.D. & Lin, C.T. 2014, 'General formulation of shift and delta operator based 2-D VLSI filter structures without global broadcast and incorporation of the symmetry', Multidimensional Systems and Signal Processing, vol. 25, no. 4, pp. 795-828.
View/Download from: Publisher's site
Having local data communication (without global broadcast of signals) among the elements is important in very large scale integration (VLSI) designs. Recently, 2-D systolic digital filter architectures were presented which eliminated the global broadcast of the input and output signals. In this paper a generalized formulation is presented that allows the derivation of various new 2-D VLSI filter structures, without global broadcast, using different 1-D filter sub-blocks and different interconnecting frameworks. The 1-D sub-blocks in z-domain are represented by general digital two-pair networks which consist of direct-form or lattice-type FIR filters in one of the frequency variables. Then, by applying the sub-blocks in various frameworks, 2-D structures realizing different transfer functions are easily obtained. As delta discrete-time operator based 1-D and 2-D digital filters (in -domain) were shown to offer better numerical accuracy and lower coefficient sensitivity in narrow-band filter designs when compared to the traditional shift-operator formulation we have covered both the conventional z-domain filters as well as delta discrete-time operator based filters. Structures realizing general 2-D IIR (both z- and -domains) and FIR transfer functions (z-domain only) are presented. As symmetry in the frequency response reduces the complexity of the design, IIR transfer functions with separable denominators, and transfer functions with quadrantal magnitude symmetry are also presented. The separable denominator frameworks are needed for quadrantal symmetry structures to guarantee BIBO stability and thus presented for both the operators. Some limitations of having exact symmetry with separable 1-D denominator factors are also discussed. &copy; 2013 Springer Science+Business Media New York.
Liao, L.D., Wu, S.L., Liou, C.H., Lu, S.W., Chen, S.A., Chen, S.F., Ko, L.W. & Lin, C.T. 2014, 'A novel 16-channel wireless system for electroencephalography measurements with dry spring-loaded sensors', IEEE Transactions on Instrumentation and Measurement, vol. 63, no. 6, pp. 1545-1555.
View/Download from: Publisher's site
Understanding brain function using electroencephalography (EEG) is an important issue for cerebral nervous system diseases, especially for epilepsy and Alzheimer's disease. Many EEG measurement systems are used reliably to study these diseases, but their bulky size and the use of wet sensors make them uncomfortable and inconvenient for users. To overcome the limitations of conventional EEG measurement systems, a wireless and wearable multichannel EEG measurement system is proposed in this paper. This system includes a wireless data acquisition device, dry spring-loaded sensors, and a sizeadjustable soft cap. We compared the performance of the proposed system using dry versus conventional wet sensors. A significant positive correlation between readings from wet and dry sensors was achieved, thus demonstrating the performance of the system. Moreover, four different features of EEG signals (i.e., normal, eye-blinking, closed-eyes, and teeth-clenching signals) were measured by 16 dry sensors to ensure that they could be detected in real-life cognitive neuroscience applications. Thus, we have shown that it is possible to reliably measure EEG signals using the proposed system. This paper presents novel insights into the field of cognitive neuroscience, showing the possibility of studying brain function under real-life conditions. &copy; 2014 IEEE.
Gramann, K., Jung, T.-.P., Ferris, D.P., Lin, C.-.T. & Makeig, S. 2014, 'Toward a new cognitive neuroscience: modeling natural brain dynamics.', Front Hum Neurosci, vol. 8, p. 444.
View/Download from: Publisher's site
Lin, C.-.T., Chuang, C.-.H., Huang, C.-.S., Tsai, S.-.F., Lu, S.-.W., Chen, Y.-.H. & Ko, L.-.W. 2014, 'Wireless and wearable EEG system for evaluating driver vigilance.', IEEE transactions on biomedical circuits and systems, vol. 8, no. 2, pp. 165-176.
View/Download from: Publisher's site
Brain activity associated with attention sustained on the task of safe driving has received considerable attention recently in many neurophysiological studies. Those investigations have also accurately estimated shifts in drivers' levels of arousal, fatigue, and vigilance, as evidenced by variations in their task performance, by evaluating electroencephalographic (EEG) changes. However, monitoring the neurophysiological activities of automobile drivers poses a major measurement challenge when using a laboratory-oriented biosensor technology. This work presents a novel dry EEG sensor based mobile wireless EEG system (referred to herein as Mindo) to monitor in real time a driver's vigilance status in order to link the fluctuation of driving performance with changes in brain activities. The proposed Mindo system incorporates the use of a wireless and wearable EEG device to record EEG signals from hairy regions of the driver conveniently. Additionally, the proposed system can process EEG recordings and translate them into the vigilance level. The study compares the system performance between different regression models. Moreover, the proposed system is implemented using JAVA programming language as a mobile application for online analysis. A case study involving 15 study participants assigned a 90 min sustained-attention driving task in an immersive virtual driving environment demonstrates the reliability of the proposed system. Consistent with previous studies, power spectral analysis results confirm that the EEG activities correlate well with the variations in vigilance. Furthermore, the proposed system demonstrated the feasibility of predicting the driver's vigilance in real time.
Lin, C.T., Lin, B.S., Lin, F.C. & Chang, C.J. 2014, 'Brain computer interface-based smart living environmental auto-adjustment control system in UPnP home networking', IEEE Systems Journal, vol. 8, no. 2, pp. 363-370.
View/Download from: Publisher's site
A brain computer interface-based smart living environmental auto-adjustment control system (BSLEACS) is proposed in this paper. Recently, many environmental control systems have been proposed to improve human quality of life. However, little research has focused on environmental control directly using the human physiological state. Based on the advantage of our technique on brain computer interface (BCI), we integrated the BCI technique with universal plug and play (UPnP) home networking for smart house applications. BSLEACS mainly consists of a wireless physiological signal acquisition module, an embedded signal processing module, a simple control protocol/power line communication environmental controller, and a host system. Here, the physiological signal acquisition module and embedded signal processing module were designed for long-term electroencephalogram (EEG) monitoring and backend analysis, respectively. The advantages of low power consumption and small volume of the above modules are suitable for smart house applications in daily life. Moreover, different from other BCI systems, the property of using only a single EEG channel to monitor cognitive state also makes BSLEACS become more practicable. BSLEACS has been verified in a practical demo room, and the environmental adjustment can be automatically controlled by the change of the user's cognitive state. BSLEACS provides a novel system prototype for environmental control, and can be simply extended and integrated with the UPnP home networking for other applications. &copy; 2012 IEEE.
Lin, Y.-.Y., Liao, S.-.H., Chang, J.-.Y. & Lin, C.-.T. 2014, 'Simplified interval type-2 fuzzy neural networks.', IEEE transactions on neural networks and learning systems, vol. 25, no. 5, pp. 959-969.
View/Download from: Publisher's site
This paper describes a self-evolving interval type-2 fuzzy neural network (FNN) for various applications. As type-1 fuzzy systems cannot effectively handle uncertainties in information within the knowledge base, we propose a simple interval type-2 FNN, which uses interval type-2 fuzzy sets in the premise and the Takagi-Sugeno-Kang (TSK) type in the consequent of the fuzzy rule. The TSK-type consequent of fuzzy rule is a linear combination of exogenous input variables. Given an initially empty the rule-base, all rules are generated with on-line type-2 fuzzy clustering. Instead of the time-consuming K-M iterative procedure, the design factors ql and qr are learned to adaptively adjust the upper and lower positions on the left and right limit outputs, using the parameter update rule based on a gradient descent algorithm. Simulation results demonstrate that our approach yields fewer test errors and less computational complexity than other type-2 FNNs.
Chuang, C.-.H., Ko, L.-.W., Jung, T.-.P. & Lin, C.-.T. 2014, 'Kinesthesia in a sustained-attention driving task.', NeuroImage, vol. 91, pp. 187-202.
View/Download from: UTS OPUS or Publisher's site
This study investigated the effects of kinesthetic stimuli on brain activities during a sustained-attention task in an immersive driving simulator. Tonic and phasic brain responses on multiple timescales were analyzed using time-frequency analysis of electroencephalographic (EEG) sources identified by independent component analysis (ICA). Sorting EEG spectra with respect to reaction times (RT) to randomly introduced lane-departure events revealed distinct effects of kinesthetic stimuli on the brain under different performance levels. Experimental results indicated that EEG spectral dynamics highly correlated with performance lapses when driving involved kinesthetic feedback. Furthermore, in the realistic environment involving both visual and kinesthetic feedback, a transitive relationship of power spectra between optimal-, suboptimal-, and poor-performance groups was found predominately across most of the independent components. In contrast to the static environment with visual input only, kinesthetic feedback reduced theta-power augmentation in the central and frontal components when preparing for action and error monitoring, while strengthening alpha suppression in the central component while steering the wheel. In terms of behavior, subjects tended to have a short response time to process unexpected events with the assistance of kinesthesia, yet only when their performance was optimal. Decrease in attentional demand, facilitated by kinesthetic feedback, eventually significantly increased the reaction time in the suboptimal-performance state. Neurophysiological evidence of mutual relationships between behavioral performance and neurocognition in complex task paradigms and experimental environments, presented in this study, might elucidate our understanding of distributed brain dynamics, supporting natural human cognition and complex coordinated, multi-joint naturalistic behavior, and lead to improved understanding of brain-behavior relations in operating enviro...
Chuang, C.-.H., Ko, L.-.W., Lin, Y.-.P., Jung, T.-.P. & Lin, C.-.T. 2014, 'Independent component ensemble of EEG for brain-computer interface.', IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society, vol. 22, no. 2, pp. 230-238.
View/Download from: Publisher's site
Recently, successful applications of independent component analysis (ICA) to electroencephalographic (EEG) signals have yielded tremendous insights into brain processes that underlie human cognition. Many studies have further established the feasibility of using independent processes to elucidate human cognitive states. However, various technical problems arise in the building of an online brain-computer interface (BCI). These include the lack of an automatic procedure for selecting independent components of interest (ICi) and the potential risk of not obtaining a desired ICi. Therefore, this study proposes an ICi-ensemble method that uses multiple classifiers with ICA processing to improve upon existing algorithms. The mechanisms that are used in this ensemble system include: 1) automatic ICi selection; 2) extraction of features of the resultant ICi; 3) the construction of parallel pipelines for effectively training multiple classifiers; and a 4) simple process that combines the multiple decisions. The proposed ICi-ensemble is demonstrated in a typical BCI application, which is the monitoring of participants' cognitive states in a realistic sustained-attention driving task. The results reveal that the proposed ICi-ensemble outperformed the previous method using a single ICi with &nbsp;&nbsp;7% (91.6% versus 84.3%) in the cognitive state classification. Additionally, the proposed ICi-ensemble method that characterizes the EEG dynamics of multiple brain areas favors the application of BCI in natural environments.
Wang, Y.K., Chen, S.A. & Lin, C.T. 2014, 'An EEG-based brain-computer interface for dual task driving detection', Neurocomputing, vol. 129, pp. 85-93.
View/Download from: Publisher's site
The development of brain-computer interfaces (BCI) for multiple applications has undergone extensive growth in recent years. Since distracted driving is a significant cause of traffic accidents, this study proposes one BCI system based on EEG for distracted driving. The removal of artifacts and the selection of useful brain sources are the essential and critical steps in the application of electroencephalography (EEG)-based BCI. In the first model, artifacts are removed, and useful brain sources are selected based on the independent component analysis (ICA). In the second model, all distracted and concentrated EEG epochs are recognized with a self-organizing map (SOM). This BCI system automatically identified independent components with artifacts for removal and detected distracted driving through the specific brain sources which are also selected automatically. The accuracy of the proposed system approached approximately 90% for the recognition of EEG epochs of distracted and concentrated driving according to the selected frontal and left motor components. &copy; 2013.
Lin, Y.Y., Chang, J.Y. & Lin, C.T. 2014, 'A TSK-type-based self-evolving compensatory interval type-2 fuzzy neural network (TSCIT2FNN) and its applications', IEEE Transactions on Industrial Electronics, vol. 61, no. 1, pp. 447-459.
View/Download from: Publisher's site
In this paper, a Takagi-Sugeno-Kang (TSK)-type-based self-evolving compensatory interval type-2 fuzzy neural network (FNN) (TSCIT2FNN) is proposed for system modeling and noise cancellation problems. A TSCIT2FNN uses type-2 fuzzy sets in an FNN in order to handle the uncertainties associated with information or data in the knowledge base. The antecedent part of each compensatory fuzzy rule is an interval type-2 fuzzy set in the TSCIT2FNN, where compensatory-based fuzzy reasoning uses adaptive fuzzy operation of a neural fuzzy system to make the fuzzy logic system effective and adaptive, and the consequent part is of the TSK type. The TSK-type consequent part is a linear combination of exogenous input variables. Initially, the rule base in the TSCIT2FNN is empty. All rules are derived according to online type-2 fuzzy clustering. For parameter learning, the consequent part parameters are tuned by a variable-expansive Kalman filter algorithm to the reinforce parameter learning ability. The antecedent type-2 fuzzy sets and compensatory weights are learned by a gradient descent algorithm to improve the learning performance. The performance of TSCIT2FNN for the identification is validated and compared with several type-1 and type-2 FNNs. Simulation results show that our approach produces smaller root-mean-square errors and converges more quickly. &copy; 1982-2012 IEEE.
Liao, S.H., Han, M.F., Chang, J.Y. & Lin, C.T. 2013, 'Study on adaptive least trimmed squares fuzzy neural network', International Journal of Fuzzy Systems, vol. 15, no. 3, pp. 338-346.
In the largest samplings of data, outliers are observations that are well separated from the major samples. To deal with outlier problems, a least trimmed squares (LTS) estimator is developed for robust linear regression problems. It is meaningful to generalize the LTS estimator to fuzzy neural network (FNN) for robust nonlinear regression problems. In addition, the determination of the trimming constant is important when using the LTS estimator. In this paper, we propose the use of an adaptive least trimmed squares fuzzy neural network (ALTS-FNN), which applies a scale estimate to a LTS-FNN. This paper particularly emphasizes the robustness of the proposed network against outliers and an automatic determination of the trimming percentage. Simulation problems are provided to compare the performance of the proposed ALTS-FNN, with an LTS-FNN and typical FNN. Simulation results show that the proposed ALTS-FNN is highly robust against outliers. &copy; 2013 TFSA.
Lin, C.-.T., Huang, K.-.C., Chuang, C.-.H., Ko, L.-.W. & Jung, T.-.P. 2013, 'Can arousing feedback rectify lapses in driving? Prediction from EEG power spectra.', Journal of neural engineering, vol. 10, no. 5, p. 056024.
View/Download from: Publisher's site
OBJECTIVE: This study explores the neurophysiological changes, measured using an electroencephalogram (EEG), in response to an arousing warning signal delivered to drowsy drivers, and predicts the efficacy of the feedback based on changes in the EEG. APPROACH: Eleven healthy subjects participated in sustained-attention driving experiments. The driving task required participants to maintain their cruising position and compensate for randomly induced lane deviations using the steering wheel, while their EEG and driving performance were continuously monitored. The arousing warning signal was delivered to participants who experienced momentary behavioral lapses, failing to respond rapidly to lane-departure events (specifically the reaction time exceeded three times the alert reaction time). MAIN RESULTS: The results of our previous studies revealed that arousing feedback immediately reversed deteriorating driving performance, which was accompanied by concurrent EEG theta- and alpha-power suppression in the bilateral occipital areas. This study further proposes a feedback efficacy assessment system to accurately estimate the efficacy of arousing warning signals delivered to drowsy participants by monitoring the changes in their EEG power spectra immediately thereafter. The classification accuracy was up 77.8% for determining the need for triggering additional warning signals. SIGNIFICANCE: The findings of this study, in conjunction with previous studies on EEG correlates of behavioral lapses, might lead to a practical closed-loop system to predict, monitor and rectify behavioral lapses of human operators in attention-critical settings.
Lin, C.-.T., Tsai, S.-.F. & Ko, L.-.W. 2013, 'EEG-based learning system for online motion sickness level estimation in a dynamic vehicle environment.', IEEE transactions on neural networks and learning systems, vol. 24, no. 10, pp. 1689-1700.
View/Download from: Publisher's site
Motion sickness is a common experience for many people. Several previous researches indicated that motion sickness has a negative effect on driving performance and sometimes leads to serious traffic accidents because of a decline in a person's ability to maintain self-control. This safety issue has motivated us to find a way to prevent vehicle accidents. Our target was to determine a set of valid motion sickness indicators that would predict the occurrence of a person's motion sickness as soon as possible. A successful method for the early detection of motion sickness will help us to construct a cognitive monitoring system. Such a monitoring system can alert people before they become sick and prevent them from being distracted by various motion sickness symptoms while driving or riding in a car. In our past researches, we investigated the physiological changes that occur during the transition of a passenger's cognitive state using electroencephalography (EEG) power spectrum analysis, and we found that the EEG power responses in the left and right motors, parietal, lateral occipital, and occipital midline brain areas were more highly correlated to subjective sickness levels than other brain areas. In this paper, we propose the use of a self-organizing neural fuzzy inference network (SONFIN) to estimate a driver's/passenger's sickness level based on EEG features that have been extracted online from five motion sickness-related brain areas, while either in real or virtual vehicle environments. The results show that our proposed learning system is capable of extracting a set of valid motion sickness indicators that originated from EEG dynamics, and through SONFIN, a neuro-fuzzy prediction model, we successfully translated the set of motion sickness indicators into motion sickness levels. The overall performance of this proposed EEG-based learning system can achieve an average prediction accuracy of ~82%.
Mcdowell, K., Lin, C.T., Oie, K.S., Jung, T.P., Gordon, S., Whitaker, K.W., Li, S.Y., Lu, S.W. & Hairston, W.D. 2013, 'Real-world neuroimaging technologies', IEEE Access, vol. 1, pp. 131-149.
View/Download from: Publisher's site
Decades of heavy investment in laboratory-based brain imaging and neuroscience have led to foundational insights into how humans sense, perceive, and interact with the external world. However, it is argued that fundamental differences between laboratory-based and naturalistic human behavior may exist. Thus, it remains unclear how well the current knowledge of human brain function translates into the highly dynamic real world. While some demonstrated successes in real-world neurotechnologies are observed, particularly in the area of brain-computer interaction technologies, innovations and developments to date are limited to a small science and technology community. We posit that advancements in realworld neuroimaging tools for use by a broad-based workforce will dramatically enhance neurotechnology applications that have the potential to radically alter human-system interactions across all aspects of everyday life. We discuss the efforts of a joint government-Academic-industry team to take an integrative, interdisciplinary, and multi-Aspect approach to translate current technologies into devices that are truly flueldable across a range of environments. Results from initial work, described here, show promise for dramatic advances in the flueld that will rapidly enhance our ability to assess brain activity in real-world scenarios. &copy; 2013 IEEE.
Lin, Y.-.Y., Chang, J.-.Y. & Lin, C.-.T. 2013, 'Identification and prediction of dynamic systems using an interactively recurrent self-evolving fuzzy neural network.', IEEE transactions on neural networks and learning systems, vol. 24, no. 2, pp. 310-321.
View/Download from: Publisher's site
This paper presents a novel recurrent fuzzy neural network, called an interactively recurrent self-evolving fuzzy neural network (IRSFNN), for prediction and identification of dynamic systems. The recurrent structure in an IRSFNN is formed as an external loops and internal feedback by feeding the rule firing strength of each rule to others rules and itself. The consequent part in the IRSFNN is composed of a Takagi-Sugeno-Kang (TSK) or functional-link-based type. The proposed IRSFNN employs a functional link neural network (FLNN) to the consequent part of fuzzy rules for promoting the mapping ability. Unlike a TSK-type fuzzy neural network, the FLNN in the consequent part is a nonlinear function of input variables. An IRSFNNs learning starts with an empty rule base and all of the rules are generated and learned online through a simultaneous structure and parameter learning. An on-line clustering algorithm is effective in generating fuzzy rules. The consequent update parameters are derived by a variable-dimensional Kalman filter algorithm. The premise and recurrent parameters are learned through a gradient descent algorithm. We test the IRSFNN for the prediction and identification of dynamic plants and compare it to other well-known recurrent FNNs. The proposed model obtains enhanced performance results.
Wu, S.-.L., Liao, L.-.D., Lu, S.-.W., Jiang, W.-.L., Chen, S.-.A. & Lin, C.-.T. 2013, 'Controlling a human-computer interface system with a novel classification method that uses electrooculography signals.', IEEE transactions on bio-medical engineering, vol. 60, no. 8, pp. 2133-2141.
View/Download from: Publisher's site
Electrooculography (EOG) signals can be used to control human-computer interface (HCI) systems, if properly classified. The ability to measure and process these signals may help HCI users to overcome many of the physical limitations and inconveniences in daily life. However, there are currently no effective multidirectional classification methods for monitoring eye movements. Here, we describe a classification method used in a wireless EOG-based HCI device for detecting eye movements in eight directions. This device includes wireless EOG signal acquisition components, wet electrodes and an EOG signal classification algorithm. The EOG classification algorithm is based on extracting features from the electrical signals corresponding to eight directions of eye movement (up, down, left, right, up-left, down-left, up-right, and down-right) and blinking. The recognition and processing of these eight different features were achieved in real-life conditions, demonstrating that this device can reliably measure the features of EOG signals. This system and its classification procedure provide an effective method for identifying eye movements. Additionally, it may be applied to study eye functions in real-life conditions in the near future.
Lin, C.T., Hsu, S.C., Chou, K.P., Siana, L. & Yang, C.T. 2013, 'Real-time boosted vehicle detection deal with high detection rate using false alarm eliminating method', International Journal of Innovative Computing, Information and Control, vol. 9, no. 7, pp. 3039-3052.
We present in this paper a high detection rate of Boosted vehicle detection. The positive database of Boosted training usually consists of similar contour or lighting of vehicle image, but in our work the training data is extracted from both daytime and evening images, it means different in lighting, and we are forced to stop Boosted training toward a high detection rate, which results in a relatively high false alarm rate. Therefore, the Boosted detector will detect vehicle candidate as much as possible. Moreover, we develop two false alarm eliminating methods to eliminate the false vehicle candidates. The algorithm consists of edge complexity for daytime case, and combination histogram matching and intensity complexity for evening case. Each case is chosen by automatic switcher algorithm. To provide real-time detection, we also proposed position based sliding window detector. The idea is based on sliding windows size selection relative to the position of vehicle candidate in an image. Thus, we do not need to apply various size of detector size as traditional Boosted algorithm. Finally, our experimental results show that our proposed system can operate in a real-world environment, while providing realtime detection. &copy; 2013 ICIC International.
Han, M.F., Liao, S.H., Chang, J.Y. & Lin, C.T. 2013, 'Dynamic group-based differential evolution using a self-adaptive strategy for global optimization problems', Applied Intelligence, vol. 39, no. 1, pp. 41-56.
View/Download from: Publisher's site
This paper describes a dynamic group-based differential evolution (GDE) algorithm for global optimization problems. The GDE algorithm provides a generalized evolution process based on two mutation operations to enhance search capability. Initially, all individuals in the population are grouped into a superior group and an inferior group based on their fitness values. The two groups perform different mutation operations. The local mutation model is applied to individuals with better fitness values, i.e., in the superior group, to search for better solutions near the current best position. The global mutation model is applied to the inferior group, which is composed of individuals with lower fitness values, to search for potential solutions. Subsequently, the GDE algorithm employs crossover and selection operations to produce offspring for the next generation. In this paper, an adaptive tuning strategy based on the well-known 1/5th rule is used to dynamically reassign the group size. It is thus helpful to trade off between the exploration ability and the exploitation ability. To validate the performance of the GDE algorithm, 13 numerical benchmark functions are tested. The simulation results indicate that the approach is effective and efficient. &copy; 2012 Springer Science+Business Media New York.
Lin, Y.Y., Chang, J.Y., Pal, N.R. & Lin, C.T. 2013, 'A mutually recurrent interval type-2 neural fuzzy system (MRIT2NFS) with self-evolving structure and parameters', IEEE Transactions on Fuzzy Systems, vol. 21, no. 3, pp. 492-509.
View/Download from: Publisher's site
In this paper, a mutually recurrent interval type-2 neural fuzzy system (MRIT2NFS) is proposed for the identification of nonlinear and time-varying systems. The MRIT2NFS uses type-2 fuzzy sets in order to enhance noise tolerance of the system. In the MRIT2NFS, the antecedent part of each recurrent fuzzy rule is defined using interval type-2 fuzzy sets, and the consequent part is of the Takagi-Sugeno-Kang type with interval weights. The antecedent part of MRIT2NFS forms a local internal feedback and interaction loop by feeding the rule firing strength of each rule to others including itself. The consequent is a linear combination of exogenous input variables. The learning of MRIT2NFS starts with an empty rule base and all rules are learned online via structure and parameter learning. The structure learning of MRIT2NFS uses online type-2 fuzzy clustering. For parameter learning, the consequent part parameters are tuned by rule-ordered Kalman filter algorithm to reinforce parameter learning ability. The type-2 fuzzy sets in the antecedent and weights representing the mutual feedback are learned by the gradient descent algorithm. After the training, a weight-elimination scheme eliminates feedback connections that do not have much effect on the network behavior. This method can efficiently remove redundant recurrence and interaction weights. Finally, the MRIT2NFS is used for system identification under both noise-free and noisy environments. For this, we consider both time series prediction and nonlinear plant modeling. Compared with type-1 recurrent fuzzy neural networks, simulation results show that our approach produces smaller root-mean-squared errors using the same number of iterations. &copy; 2013 IEEE.
Han, M.F., Lin, C.T. & Chang, J.Y. 2013, 'Efficient differential evolution algorithm-based optimisation of fuzzy prediction model for time series forecasting', International Journal of Intelligent Information and Database Systems, vol. 7, no. 3, pp. 225-241.
View/Download from: Publisher's site
This paper proposes a differential evolution algorithm with efficient mutation strategy (DEEMS) for fuzzy prediction model (FPM) optimisation. The proposed DEEMS uses a modified mutation operation which considers local information nearby each individual to trade-off between the exploration ability and the exploitation ability. In the FPM design, we adopt an entropy measure method to determine the number of rules. Initially, there is no rule in the FPM. Fuzzy rules are automatically generated by entropy measure. Subsequently, the DEEMS algorithm is performed to optimise all the free parameters. During evolution process, the scale factor and crossover rate in the DEEMS algorithm are adjusted by adaptive parameter tuning strategy for each generation. It is thus helpful to enhance the robustness of the DEEMS algorithm. In the simulation, the proposed FPM with DEEMS model (FPM-DEEMS) is applied to two real world problems. Results show that the proposed FPM-DEEMS model obtains better performance than other algorithms. Copyright &copy; 2013 Inderscience Enterprises Ltd.
Li, S.Y., Yang, C.H., Ko, L.W., Lin, C.T. & Ge, Z.M. 2013, 'Implementation on electronic circuits and RTR pragmatical adaptive synchronization: Time-reversed uncertain dynamical systems' analysis and applications', Abstract and Applied Analysis, vol. 2013.
View/Download from: Publisher's site
We expose the chaotic attractors of time-reversed nonlinear system, further implement its behavior on electronic circuit, and apply the pragmatical asymptotically stability theory to strictly prove that the adaptive synchronization of given master and slave systems with uncertain parameters can be achieved. In this paper, the variety chaotic motions of time-reversed Lorentz system are investigated through Lyapunov exponents, phase portraits, and bifurcation diagrams. For further applying the complex signal in secure communication and file encryption, we construct the circuit to show the similar chaotic signal of time-reversed Lorentz system. In addition, pragmatical asymptotically stability theorem and an assumption of equal probability for ergodic initial conditions (Ge et al., 1999, Ge and Yu, 2000, and Matsushima, 1972) are proposed to strictly prove that adaptive control can be accomplished successfully. The current scheme of adaptive control - by traditional Lyapunov stability theorem and Barbalat lemma, which are used to prove the error vector - approaches zero, as time approaches infinity. However, the core question - why the estimated or given parameters also approach to the uncertain parameters - remains without answer. By the new stability theory, those estimated parameters can be proved approaching the uncertain values strictly, and the simulation results are shown in this paper. &copy; 2013 Shih-Yu Li et al.
Han, M.F., Lin, C.T. & Chang, J.Y. 2013, 'Differential evolution with local information for neuro-fuzzy systems optimisation', Knowledge-Based Systems, vol. 44, pp. 78-89.
View/Download from: Publisher's site
This paper proposes a differential evolution with local information (DELI) algorithm for Takagi-Sugeno-Kang-type (TSK-type) neuro-fuzzy systems (NFSs) optimisation. The DELI algorithm uses a modified mutation operation that considers a neighbourhood relationship for each individual to maintain the diversity of the population and to increase the search capability. This paper also proposes an adaptive fuzzy c-means method for determining the number of rules and for identifying suitable initial parameters for the rules. Initially, there are no rules in the NFS model; the rules are automatically generated by the fuzzy measure and the fuzzy c-means method. Until the firing strengths of all of the training patterns satisfy a pre-specified threshold, the process of rule generation is terminated. Subsequently, the DELI algorithm optimises all of the free parameters for NFSs design. To enhance the performance of the DELI algorithm, an adaptive parameter tuning based on the 1/5th rule is used for the tuning scale factor F. The 1/5th rule dynamically adjusts the tuning scale factor in each period to enhance the search capability of the DELI algorithm. Finally, the proposed NFS with DELI model (NFS-DELI) is applied to nonlinear control and prediction problems. The results of this paper demonstrate the effectiveness of the proposed NFS-DELI model. &copy; 2013 Elsevier B.V. All rights reserved.
Lin, C.T. 2013, 'ABC intelligence on fuzziness', Studies in Fuzziness and Soft Computing, vol. 298, pp. 377-381.
View/Download from: Publisher's site
Li, S.Y., Yang, C.H., Lin, C.T., Ko, L.W. & Chiu, T.T. 2013, 'Chaotic motions in the real fuzzy electronic circuits', Abstract and Applied Analysis, vol. 2013.
View/Download from: Publisher's site
Fuzzy electronic circuit (FEC) is firstly introduced, which is implementing Takagi-Sugeno (T-S) fuzzy chaotic systems on electronic circuit. In the research field of secure communications, the original source should be blended with other complex signals. Chaotic signals are one of the good sources to be applied to encrypt high confidential signals, because of its high complexity, sensitiveness of initial conditions, and unpredictability. Consequently, generating chaotic signals on electronic circuit to produce real electrical signals applied to secure communications is an exceedingly important issue. However, nonlinear systems are always composed of many complex equations and are hard to realize on electronic circuits. Takagi-Sugeno (T-S) fuzzy model is a powerful tool, which is described by fuzzy IF-THEN rules to express the local dynamics of each fuzzy rule by a linear system model. Accordingly, in this paper, we produce the chaotic signals via electronic circuits through T-S fuzzy model and the numerical simulation results provided by MATLAB are also proposed for comparison. T-S fuzzy chaotic Lorenz and Chen-Lee systems are used for examples and are given to demonstrate the effectiveness of the proposed electronic circuit. &copy; 2013 Shih-Yu Li et al.
Han, M.F., Lin, C.T., Chang, J.Y. & Li, D.L. 2013, 'Group-based differential evolution for numerical optimization problems', International Journal of Innovative Computing, Information and Control, vol. 9, no. 3, pp. 1357-1372.
This paper proposes a group-based differential evolution (GDE) algorithm for numerical optimization problems. The proposed GDE algorithm provides a new process using two mutation strategies to effectively enhance the search for the globally optimal solution. Initially, all individuals in the population are partitioned into an elite group and an inferior group based on their fitness value. In the elite group, individuals with a better fitness value employ the local mutation operation to search for better solutions near the current best individual. The inferior group, which is composed of individuals with worse fitness values, uses a global mutation operation to search for potential solutions and to increase the diversity of the population. Subsequently, the GDE algorithm employs crossover and selection operations to produce offspring for the next generation. This paper also proposes two parameter-tuning strategies for the robustness of the GDE algorithm in the evolution process. To validate the performance of the GDE algorithm, 13 well-known numerical benchmark functions were tested on low- and high-dimensional problems. The simulation results indicate that our approach is efficient. &copy; 2013 ICIC International.
Lin, C.-.L., Jung, T.-.P., Chuang, S.-.W., Duann, J.-.R., Lin, C.-.T. & Chiu, T.-.W. 2013, 'Self-adjustments may account for the contradictory correlations between HRV and motion-sickness severity.', International journal of psychophysiology : official journal of the International Organization of Psychophysiology, vol. 87, no. 1, pp. 70-80.
View/Download from: Publisher's site
This study investigates the relationship between heart rate variability (HRV) and the level of motion sickness (MS) induced by simulated tunnel driving. The HRV indices, normalized low frequency (NLF, 0.04-0.15 Hz), normalized high frequency (NHF, 0.15-0.4 Hz), and LF/HF ratio were correlated with the subjectively and continuously rated MS levels of 20 participants. The experimental results showed that for 13 of the subjects, the MS levels positively correlated with the NLF and the LF/HF ratio and negatively correlated with the NHF. The remaining seven subjects had negative correlations between the MS levels and the NLF and the LF/HF ratio and a positive correlation between the MS levels and the NHF. To clarify this contradiction, this study also inspected the effects of subjects' self-adjustments on the correlations between the MS levels and the HRV indices and showed that the variations in the relationship might be attributed to the subjects' self-adjustments, which they used to relieve the discomfort of MS.
Li, S.Y., Yang, C.H., Chen, S.A., Ko, L.W. & Lin, C.T. 2013, 'Fuzzy adaptive synchronization of time-reversed chaotic systems via a new adaptive control strategy', Information Sciences, vol. 222, pp. 486-500.
View/Download from: Publisher's site
A novel adaptive control strategy is proposed herein to increase the efficiency of adaptive control by combining Takagi-Sugeno (T-S) fuzzy modeling and the Ge-Yao-Chen (GYC) partial region stability theory. This approach provides two major contributions: (1) increased synchronization efficiency, especially for parameters tracing and (2) a simpler controller design. Two simulated cases are presented for comparison: Case 1 utilizes normal adaptive synchronization, whereas Case 2 utilizes the Takagi-Sugeno (T-S) fuzzy model-based Lorenz systems to realize adaptive synchronization via the new adaptive scheme. The simulation results demonstrate the effectiveness and feasibility of our new adaptive strategy. &copy; 2012 Elsevier Inc. All rights reserved.
Wu, S.-.L., Liao, L.-.D., Liou, C.-.H., Chen, S.-.A., Ko, L.-.W., Chen, B.-.W., Wang, P.-.S., Chen, S.-.F. & Lin, C.-.T. 2012, 'Design of the multi-channel electroencephalography-based brain-computer interface with novel dry sensors.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, vol. 2012, pp. 1793-1797.
View/Download from: Publisher's site
The traditional brain-computer interface (BCI) system measures the electroencephalography (EEG) signals by the wet sensors with the conductive gel and skin preparation processes. To overcome the limitations of traditional BCI system with conventional wet sensors, a wireless and wearable multi-channel EEG-based BCI system is proposed in this study, including the wireless EEG data acquisition device, dry spring-loaded sensors, a size-adjustable soft cap. The dry spring-loaded sensors are made of metal conductors, which can measure the EEG signals without skin preparation and conductive gel. In addition, the proposed system provides a size-adjustable soft cap that can be used to fit user's head properly. Indeed, the results are shown that the proposed system can properly and effectively measure the EEG signals with the developed cap and sensors, even under movement. In words, the developed wireless and wearable BCI system is able to be used in cognitive neuroscience applications.
Li, D.L., Prasad, M., Hsu, S.C., Hong, C.T. & Lin, C.T. 2012, 'Face recognition using nonparametric-weighted Fisherfaces', Eurasip Journal on Advances in Signal Processing, vol. 2012, no. 1.
View/Download from: Publisher's site
This study presents an appearance-based face recognition scheme called the nonparametric-weighted Fisherfaces (NW-Fisherfaces). Pixels in a facial image are considered as coordinates in a high-dimensional space and are transformed into a face subspace for analysis by using nonparametric-weighted feature extraction (NWFE). According to previous studies of hyperspectral image classification, NWFE is a powerful tool for extracting hyperspectral image features. The Fisherfaces method maximizes the ratio of between-class scatter to that of within-class scatter. In this study, the proposed NW-Fisherfaces weighted the between-class scatter to emphasize the boundary structure of the transformed face subspace and, therefore, enhances the separability for different persons' face. The proposed NW-Fisherfaces was compared with Orthogonal Laplacianfaces, Eigenfaces, Fisherfaces, direct linear discriminant analysis, and null space linear discriminant analysis methods for tests on five facial databases. Experimental results showed that the proposed approach outperforms other feature extraction methods for most databases. &copy; 2012 Li et al.
Li, S.Y., Yang, C.H., Lin, C.T., Ko, L.W. & Chiu, T.T. 2012, 'Adaptive synchronization of chaotic systems with unknown parameters via new backstepping strategy', Nonlinear Dynamics, vol. 70, no. 3, pp. 2129-2143.
View/Download from: Publisher's site
In this paper, a new effective approach - backstepping with Ge-Yao-Chen (GYC) partial region stability theory (called BGYC in this article) is proposed to applied to adaptive synchronization. Backstepping design is a recursive procedure that combines the choice of a Lyapunov function with the design of a controller, and it presents a systematic procedure for selecting a proper controller in chaos synchronization. We further combine the systematic backstepping design and GYC partial region stability theory in this article, Lyapunov function can be chosen as a simple linear homogeneous function of states, and the controllers and the update laws of parameters shall be much simpler. Further, it also introduces less simulation error - the numerical simulation results show that the states errors and parametric errors approach to zero much more exactly and efficiently, which are compared with the original one. Two cases are presented in the simulation results to show the effectiveness and feasibility of our new strategy. &copy; Springer Science+Business Media B.V. 2012.
Lin, C.T., Li, D.L., Lai, J.H., Han, M.F. & Chang, J.Y. 2012, 'Automatic age estimation system for face images', International Journal of Advanced Robotic Systems, vol. 9.
View/Download from: Publisher's site
Humans are the most important tracking objects in surveillance systems. However, human tracking is not enough to provide the required information for personalized recognition. In this paper, we present a novel and reliable framework for automatic age estimation based on computer vision. It exploits global face features based on the combination of Gabor wavelets and orthogonal locality preserving projections. In addition, the proposed system can extract face aging features automatically in real-time. This means that the proposed system has more potential in applications compared to other semi-automatic systems. The results obtained from this novel approach could provide clearer insight for operators in the field of age estimation to develop real-world applications. &copy; 2012 Lin et al.
Liao, L.-.D., Lin, C.-.T., Shih, Y.-.Y.I., Lai, H.-.Y., Zhao, W.-.T., Duong, T.Q., Chang, J.-.Y., Chen, Y.-.Y. & Li, M.-.L. 2012, 'Investigation of the cerebral hemodynamic response function in single blood vessels by functional photoacoustic microscopy.', Journal of biomedical optics, vol. 17, no. 6, p. 061210.
View/Download from: Publisher's site
The specificity of the hemodynamic response function (HRF) is determined spatially by the vascular architecture and temporally by the evolution of hemodynamic changes. Here, we used functional photoacoustic microscopy (fPAM) to investigate single cerebral blood vessels of rats after left forepaw stimulation. In this system, we analyzed the spatiotemporal evolution of the HRFs of the total hemoglobin concentration (HbT), cerebral blood volume (CBV), and hemoglobin oxygen saturation (SO(2)). Changes in specific cerebral vessels corresponding to various electrical stimulation intensities and durations were bilaterally imaged with 36 65-m(2) spatial resolution. Stimulation intensities of 1, 2, 6, and 10 mA were applied for periods of 5 or 15 s. Our results show that the relative functional changes in HbT, CBV, and SO(2) are highly dependent not only on the intensity of the stimulation, but also on its duration. Additionally, the duration of the stimulation has a strong influence on the spatiotemporal characteristics of the HRF as shorter stimuli elicit responses only in the local vasculature (smaller arterioles), whereas longer stimuli lead to greater vascular supply and drainage. This study suggests that the current fPAM system is reliable for studying relative cerebral hemodynamic changes, as well as for offering new insights into the dynamics of functional cerebral hemodynamic changes in small animals.
Lin, F.C., Ko, L.W., Chuang, C.H., Su, T.P. & Lin, C.T. 2012, 'Generalized EEG-based drowsiness prediction system by using a self-organizing neural fuzzy system', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 59, no. 9, pp. 2044-2055.
View/Download from: Publisher's site
A generalized EEG-based Neural Fuzzy system to predict driver's drowsiness was proposed in this study. Driver's drowsy state monitoring system has been implicated as a causal factor for the safety driving issue, especially when the driver fell asleep or distracted in driving. However, the difficulties in developing such a system are lack of significant index for detecting the driver's drowsy state in real-time and the interference of the complicated noise in a realistic and dynamic driving environment. In our past studies, we found that the electroencephalogram (EEG) power spectrum changes were highly correlated with the driver's behavior performance especially the occipital component. Different from presented subject-dependent drowsy state monitor systems, whose system performance may decrease rapidly when different subject applies with the drowsiness detection model constructed by others, in this study, we proposed a generalized EEG-based Self-organizing Neural Fuzzy system to monitor and predict the driver's drowsy state with the occipital area. Two drowsiness prediction models, subject-dependent and generalized cross-subject predictors, were investigated in this study for system performance analysis. Correlation coefficients and root mean square errors are showed as the experimental results and interpreted the performances of the proposed system significantly better than using other traditional Neural Networks (p-value <0.038). Besides, the proposed EEG-based Self-organizing Neural Fuzzy system can be generalized and applied in the subjects' independent sessions. This unique advantage can be widely used in the real-life applications. &copy; 2012 IEEE.
Chuang, S.-.W., Ko, L.-.W., Lin, Y.-.P., Huang, R.-.S., Jung, T.-.P. & Lin, C.-.T. 2012, 'Co-modulatory spectral changes in independent brain processes are correlated with task performance.', NeuroImage, vol. 62, no. 3, pp. 1469-1477.
View/Download from: Publisher's site
This study investigates the independent modulators that mediate the power spectra of electrophysiological processes, measured by electroencephalogram (EEG), in a sustained-attention experiment. EEG and behavioral data were collected during 1-2 hour virtual-reality based driving experiments in which subjects were instructed to maintain their cruising position and compensate for randomly induced drift using the steering wheel. Independent component analysis (ICA) applied to 30-channel EEG data separated the recorded EEG signals into a sum of maximally temporally independent components (ICs) for each of 30 subjects. Logarithmic spectra of resultant IC activities were then decomposed by principal component analysis, followed by ICA, to find spectrally fixed and temporally independent modulators (IM). Across subjects, the spectral ICA consistently found four performance-related independent modulators: delta, delta-theta, alpha, and beta modulators that multiplicatively affected the spectra of spatially distinct IC processes when the participants experienced waves of alternating alertness and drowsiness during long-hour simulated driving. The activation of the delta-theta modulator increased monotonically as subjects' task performances decreased. Furthermore, the time courses of the theta-beta modulator were highly correlated with concurrent changes in driving errors across subjects (r=0.77&plusmn;0.13).
Lin, C.T., Huang, T.Y., Lin, W.J., Chang, S.Y., Lin, Y.H., Ko, L.W., Hung, D.L. & Chang, E.C. 2012, 'Gender differences in wayfinding in virtual environments with global or local landmarks', Journal of Environmental Psychology, vol. 32, no. 2, pp. 89-96.
View/Download from: Publisher's site
This study assesses gender differences in wayfinding in environments with global or local landmarks by analyzing both overall and fine-grained measures of performance. Both female and male participants were required to locate targets in grid-like virtual environments with local or global landmarks. Interestingly, the results of the two overall measures did not converge: although females spent more time than males in locating targets, both genders were generally equivalent in terms of corrected travel path. Fine-grained measures account for different aspects of wayfinding behavior and provide additional information that explains the divergence in overall measures; females spent less time traveling away from the target location, a higher proportion of time not traversing, and made more rotations when stopping than males did. Rather than unequivocally supporting male superiority in wayfinding tasks, both the overall and fine-grained measures partially indicate that males and females are differentially superior when using global and local landmark information, respectively. To summarize, males moved faster than females but did not necessarily navigate the spatial surroundings more efficiently. Each gender showed different strengths related to wayfinding; these differences require the application of both overall and fine-grained measures for accurate assessment. &copy; 2012 Elsevier Ltd.
Su, M.T., Lin, C.T. & Hsu, K.W. 2012, 'A novel method for locating solder joints based on modified binary potential function', International Journal of Innovative Computing, Information and Control, vol. 8, no. 1 B, pp. 911-932.
The solder joint location on inductors is an extremely important aspect of the industrial process; in particular, the visual location plays a fundamental role. The visual-location process is often carried out by human experts. The disadvantages associated with manual location are the large amount of time required and reduced efficiency as operator fatigue occurs. In addition, as electronic products now tend to be miniaturized, portable and with dense component layout, manual location is becoming unreliable. This has prompted the development of automatic visual-location systems to speed up the location process, increase production efficiency and improve manufacture yield rate. In this paper, vie propose an automatic visual-location method for solder joints to address the problem of feature extraction in digital images, using the concept of potential functions (PF). In order to make the location method more suitable for the inductor industry, the virtual external electric field concept is introduced. The proposed location system, which uses modified binary potential functions (MBPF), has been implemented and tested with three kinds of inductors. The experimental results show that the proposed scheme performs with a high degree of accuracy, even with testing samples that are significantly different in appearance. &copy; 2012 ICIC International.
Li, C.H., Ho, H.H., Liu, Y.L., Lin, C.T., Kuo, B.C. & Taur, J.S. 2012, 'An automatic method for selecting the parameter of the normalized kernel function to support vector machines', Journal of Information Science and Engineering, vol. 28, no. 1, pp. 1-15.
Soft-margin support vector machine (SVM) is one of the most powerful techniques for supervised classification. However, the performances of SVMs are based on choosing the proper kernel functions or proper parameters of a kernel function. It is extremely time consuming by applying the k-fold cross-validation (CV) to choose the almost best parameter. Nevertheless, the searching range and fineness of the grid method should be determined in advance. In this paper, an automatic method for selecting the parameter of the normalized kernel function is proposed. In the experimental results, it costs very little time than k-fold cross-validation for selecting the parameter by our proposed method. Moreover, the corresponding soft-margin SVMs can obtain more accurate or at least equal performance than the soft-margin SVMs by applying k-fold cross-validation to determine the parameters.
Lin, C.-.L., Shaw, F.-.Z., Young, K.-.Y., Lin, C.-.T. & Jung, T.-.P. 2012, 'EEG correlates of haptic feedback in a visuomotor tracking task.', NeuroImage, vol. 60, no. 4, pp. 2258-2273.
View/Download from: Publisher's site
This study investigates the temporal brain dynamics associated with haptic feedback in a visuomotor tracking task. Haptic feedback with deviation-related forces was used throughout tracking experiments in which subjects' behavioral responses and electroencephalogram (EEG) data were simultaneously measured. Independent component analysis was employed to decompose the acquired EEG signals into temporally independent time courses arising from distinct brain sources. Clustering analysis was used to extract independent components that were comparable across participants. The resultant independent brain processes were further analyzed via time-frequency analysis (event-related spectral perturbation) and event-related coherence (ERCOH) to contrast brain activity during tracking experiments with or without haptic feedback. Across subjects, in epochs with haptic feedback, components with equivalent dipoles in or near the right motor region exhibited greater alpha band power suppression. Components with equivalent dipoles in or near the left frontal, central, left motor, right motor, and parietal regions exhibited greater beta-band power suppression, while components with equivalent dipoles in or near the left frontal, left motor, and right motor regions showed greater gamma-band power suppression relative to non-haptic conditions. In contrast, the right occipital component cluster exhibited less beta-band power suppression in epochs with haptic feedback compared to non-haptic conditions. The results of ERCOH analysis of the six component clusters showed that there were significant increases in coherence between different brain networks in response to haptic feedback relative to the coherence observed when haptic feedback was not present. The results of this study provide novel insight into the effects of haptic feedback on the brain and may aid the development of new tools to facilitate the learning of motor skills.
Lee, C.Y., Lin, C.T., Hong, C.T. & Su, M.T. 2012, 'Smoke detection using spatial and temporal analyses', International Journal of Innovative Computing, Information and Control, vol. 8, no. 7 A, pp. 4749-4770.
Video-based fire detection is currently a fairly common application with the growth in the number of installed surveillance video systems. Moreover, the related processing units are becoming more powerful. Smoke is an early sign of most fires; therefore, selecting an appropriate smoke-detection method is essential. However, detecting smoke without creating a false alarm remains a challenging problem for open or large spaces with the disturbances of common moving objects, such as pedestrians and vehicles. This study proposes a novel video-based smoke-detection method that can be incorpora,ted into a surveillance system to provide early alerts. In this study, the process of extracting smoke features from Candidate regions was accomplished by analyzing the spatial and temporal characteristics of video sequences for three important features: edge blurring, gradual energy changes, and gradual chromatic configuration changes. The proposed spatial-temporal analysis technique improves the feature extraction of gradual energy changes. In order to make the video smoke-detection results more reliable, these three features were combined using a support vector machine (SVM) technique and a temporal-based alarm decision unit (ADU) was also introduced. The effectiveness of the proposed algorithm was evaluated on a PC with an Intel&reg; Core 2 Duo CPU (2.2 GHz) and 2 GB RAM. The average processing time was 32.27 ms per frame; i.e., the proposed algorithm can process 30.98 frames per second. Experimental results showed that the proposed system can detect smoke effectively with a low false-alarm rate and a short reaction time in many real-world scenarios. &copy; ICIC International 2012.
Liao, L.-.D., Lin, C.-.T., Shih, Y.-.Y.I., Duong, T.Q., Lai, H.-.Y., Wang, P.-.H., Wu, R., Tsang, S., Chang, J.-.Y., Li, M.-.L. & Chen, Y.-.Y. 2012, 'Transcranial imaging of functional cerebral hemodynamic changes in single blood vessels using in vivo photoacoustic microscopy.', Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism, vol. 32, no. 6, pp. 938-951.
View/Download from: Publisher's site
Optical imaging of changes in total hemoglobin concentration (HbT), cerebral blood volume (CBV), and hemoglobin oxygen saturation (SO(2)) provides a means to investigate brain hemodynamic regulation. However, high-resolution transcranial imaging remains challenging. In this study, we applied a novel functional photoacoustic microscopy technique to probe the responses of single cortical vessels to left forepaw electrical stimulation in mice with intact skulls. Functional changes in HbT, CBV, and SO(2) in the superior sagittal sinus and different-sized arterioles from the anterior cerebral artery system were bilaterally imaged with unambiguous 36 65-m(2) spatial resolution. In addition, an early decrease of SO(2) in single blood vessels during activation (i.e., 'the initial dip') was observed. Our results indicate that the initial dip occurred specifically in small arterioles of activated regions but not in large veins. This technique complements other existing imaging approaches for the investigation of the hemodynamic responses in single cerebral blood vessels.
Su, M.T., Lin, C.T., Hsu, S.C., Li, D.L., Lin, C.J. & Chen, C.H. 2012, 'Nonlinear system control using functional-link-based neuro-fuzzy network model embedded with modified particle swarm optimizer', International Journal of Fuzzy Systems, vol. 14, no. 1, pp. 97-109.
This study presents an evolutionary neural fuzzy system (NFS) for nonlinear system control. The proposed NFS model uses functional link neural networks (FLNNs) as the consequent part of the fuzzy rules. This study uses orthogonal polynomials and linearly independent functions in a functional expansion of the functional link neural networks. A learning algorithm, which consists of structure learning and parameter learning, is presented. The structure learning depends on the entropy measure to determine the number of fuzzy rules. The parameter learning, based on the particle swarm optimization (PSO) algorithm, can adjust the shape of the membership function and the corresponding weighting of the FLNN. The distance-based mutation operator, which strongly encourages a global search giving the particles more chance of converging to the global optimum, is introduced. The simulation results have shown the proposed method can improve the searching ability and is very suitable for the nonlinear system control applications. &copy; 2012 TFSA.
Lai, H.-.Y., Liao, L.-.D., Lin, C.-.T., Hsu, J.-.H., He, X., Chen, Y.-.Y., Chang, J.-.Y., Chen, H.-.F., Tsang, S. & Shih, Y.-.Y.I. 2012, 'Design, simulation and experimental validation of a novel flexible neural probe for deep brain stimulation and multichannel recording.', Journal of neural engineering, vol. 9, no. 3, p. 036001.
View/Download from: Publisher's site
An implantable micromachined neural probe with multichannel electrode arrays for both neural signal recording and electrical stimulation was designed, simulated and experimentally validated for deep brain stimulation (DBS) applications. The developed probe has a rough three-dimensional microstructure on the electrode surface to maximize the electrode-tissue contact area. The flexible, polyimide-based microelectrode arrays were each composed of a long shaft (14.9 mm in length) and 16 electrodes (5 &micro;m thick and with a diameter of 16 &micro;m). The ability of these arrays to record and stimulate specific areas in a rat brain was evaluated. Moreover, we have developed a finite element model (FEM) applied to an electric field to evaluate the volume of tissue activated (VTA) by DBS as a function of the stimulation parameters. The signal-to-noise ratio ranged from 4.4 to 5 over a 50 day recording period, indicating that the laboratory-designed neural probe is reliable and may be used successfully for long-term recordings. The somatosensory evoked potential (SSEP) obtained by thalamic stimulations and in vivo electrode-electrolyte interface impedance measurements was stable for 50 days and demonstrated that the neural probe is feasible for long-term stimulation. A strongly linear (positive correlation) relationship was observed among the simulated VTA, the absolute value of the SSEP during the 200 ms post-stimulus period (SSEP) and c-Fos expression, indicating that the simulated VTA has perfect sensitivity to predict the evoked responses (c-Fos expression). This laboratory-designed neural probe and its FEM simulation represent a simple, functionally effective technique for studying DBS and neural recordings in animal models.
Liao, L.-.D., Chen, C.-.Y., Wang, I.-.J., Chen, S.-.F., Li, S.-.Y., Chen, B.-.W., Chang, J.-.Y. & Lin, C.-.T. 2012, 'Gaming control using a wearable and wireless EEG-based brain-computer interface device with novel dry foam-based sensors.', Journal of neuroengineering and rehabilitation, vol. 9, p. 5.
View/Download from: Publisher's site
A brain-computer interface (BCI) is a communication system that can help users interact with the outside environment by translating brain signals into machine commands. The use of electroencephalographic (EEG) signals has become the most common approach for a BCI because of their usability and strong reliability. Many EEG-based BCI devices have been developed with traditional wet- or micro-electro-mechanical-system (MEMS)-type EEG sensors. However, those traditional sensors have uncomfortable disadvantage and require conductive gel and skin preparation on the part of the user. Therefore, acquiring the EEG signals in a comfortable and convenient manner is an important factor that should be incorporated into a novel BCI device. In the present study, a wearable, wireless and portable EEG-based BCI device with dry foam-based EEG sensors was developed and was demonstrated using a gaming control application. The dry EEG sensors operated without conductive gel; however, they were able to provide good conductivity and were able to acquire EEG signals effectively by adapting to irregular skin surfaces and by maintaining proper skin-sensor impedance on the forehead site. We have also demonstrated a real-time cognitive stage detection application of gaming control using the proposed portable device. The results of the present study indicate that using this portable EEG-based BCI device to conveniently and effectively control the outside world provides an approach for researching rehabilitation engineering.
Chiu, T.-.C., Gramann, K., Ko, L.-.W., Duann, J.-.R., Jung, T.-.P. & Lin, C.-.T. 2012, 'Alpha modulation in parietal and retrosplenial cortex correlates with navigation performance.', Psychophysiology, vol. 49, no. 1, pp. 43-55.
View/Download from: Publisher's site
The present study investigated the brain dynamics accompanying spatial navigation based on distinct reference frames. Participants preferentially using an allocentric or an egocentric reference frame navigated through virtual tunnels and reported their homing direction at the end of each trial based on their spatial representation of the passage. Task-related electroencephalographic (EEG) dynamics were analyzed based on independent component analysis (ICA) and subsequent clustering of independent components. Parietal alpha desynchronization during encoding of spatial information predicted homing performance for participants using an egocentric reference frame. In contrast, retrosplenial and occipital alpha desynchronization during retrieval covaried with homing performance of participants using an allocentric reference frame. These results support the assumption of distinct neural networks underlying the computation of distinct reference frames and reveal a direct relationship of alpha modulation in parietal and retrosplenial areas with encoding and retrieval of spatial information for homing behavior.
Lin, Y.-.P., Chen, J.-.H., Duann, J.-.R., Lin, C.-.T. & Jung, T.-.P. 2011, 'Generalizations of the subject-independent feature set for music-induced emotion recognition.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, vol. 2011, pp. 6092-6095.
View/Download from: Publisher's site
Electroencephalogram (EEG)-based emotion recognition has been an intensely growing field. Yet, how to achieve acceptable accuracy on a practical system with as fewer electrodes as possible is less concerned. This study evaluates a set of subject-independent features, based on differential power asymmetry of symmetric electrode pairs [1], with emphasis on its applicability to subject variability in music-induced emotion classification problem. Results of this study have evidently validated the feasibility of using subject-independent EEG features to classify four emotional states with acceptable accuracy in second-scale temporal resolution. These features could be generalized across subjects to detect emotion induced by music excerpts not limited to the music database that was used to derive the emotion-specific features.
Lin, C.-.T., Lin, C.-.L., Chiu, T.-.W., Duann, J.-.R. & Jung, T.-.P. 2011, 'Effect of respiratory modulation on relationship between heart rate variability and motion sickness.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, vol. 2011, pp. 1921-1924.
View/Download from: Publisher's site
This study investigates the interplay among heart rate variability (HRV), respiration, and the severity of motion sickness (MS) in a realistic passive driving task. Although HRV is a commonly used metrically in physiological research or even believed to be a direct measure of sympathovagal activities, the results of MS-effected HRV remain mixed across studies. The goal of this study is to find the source of these contradicting results of HRV associated with MS. Experimental results of this study showed that the group trend of the low-frequency (LF) component and the LF/HF ratio increased and high-frequency (HF) component decreased significantly as self-reported MS level increased (p<0.001), consistent with a perception-driven autonomic response of the cardiovascular system. However, in one of the subjects, the relationship was reversed when individuals intentionally adjust themselves (deep breathing) to relieve the discomfort of MS during the experiments. It appears that the correlations between HRV and MS level were higher when individuals made fewer adjustments (the number of deep breathing) during the passive driving experiments.
Lin, C.J., Chen, C.H. & Lin, C.T. 2011, 'An efficient evolutionary algorithm for fuzzy inference systems', Evolving Systems, vol. 2, no. 2, pp. 83-99.
View/Download from: Publisher's site
In this paper, a novel self-constructing evolutionary algorithm (SCEA) for designing a TSK-type fuzzy model (TFM) is proposed. The proposed SCEA method is different from normal genetic algorithms (GAs). A chromosome of a population in traditional GAs represents a full solution and only one population presents all solutions in each generation. Our proposed method uses a population to evaluate a partial solution locally and applies several populations to construct a full solution. Thus, a chromosome represents only a partial solution. The proposed SCEA method uses the self-constructing learning algorithm to construct the TFM automatically that is based on the input training data to decide on the input partition. Fuzzy rules are created and begin to grow as the first training pattern arrives. Thus, the user need not give any a priori knowledge or even any initial information on the SCEA. We also adopted the sequence search-based dynamic evolution (SSDE) method to carry out parameter learning of the TFM. Simulation results have shown that the proposed SCEA method performs better than some existing methods. &copy; 2010 Springer-Verlag.
Chen, S.-.W., Lin, S.-.H., Liao, L.-.D., Lai, H.-.Y., Pei, Y.-.C., Kuo, T.-.S., Lin, C.-.T., Chang, J.-.Y., Chen, Y.-.Y., Lo, Y.-.C., Chen, S.-.Y., Wu, R. & Tsang, S. 2011, 'Quantification and recognition of parkinsonian gait from monocular video imaging using kernel-based principal component analysis.', Biomedical engineering online, vol. 10, p. 99.
View/Download from: Publisher's site
BACKGROUND: The computer-aided identification of specific gait patterns is an important issue in the assessment of Parkinson's disease (PD). In this study, a computer vision-based gait analysis approach is developed to assist the clinical assessments of PD with kernel-based principal component analysis (KPCA). METHOD: Twelve PD patients and twelve healthy adults with no neurological history or motor disorders within the past six months were recruited and separated according to their "Non-PD", "Drug-On", and "Drug-Off" states. The participants were asked to wear light-colored clothing and perform three walking trials through a corridor decorated with a navy curtain at their natural pace. The participants' gait performance during the steady-state walking period was captured by a digital camera for gait analysis. The collected walking image frames were then transformed into binary silhouettes for noise reduction and compression. Using the developed KPCA-based method, the features within the binary silhouettes can be extracted to quantitatively determine the gait cycle time, stride length, walking velocity, and cadence. RESULTS AND DISCUSSION: The KPCA-based method uses a feature-extraction approach, which was verified to be more effective than traditional image area and principal component analysis (PCA) approaches in classifying "Non-PD" controls and "Drug-Off/On" PD patients. Encouragingly, this method has a high accuracy rate, 80.51%, for recognizing different gaits. Quantitative gait parameters are obtained, and the power spectrums of the patients' gaits are analyzed. We show that that the slow and irregular actions of PD patients during walking tend to transfer some of the power from the main lobe frequency to a lower frequency band. Our results indicate the feasibility of using gait performance to evaluate the motor function of patients with PD. CONCLUSION: This KPCA-based method requires only a digital camera and a decorated corridor setup. The ease of use a...
Gramann, K., Gwin, J.T., Ferris, D.P., Oie, K., Jung, T.-.P., Lin, C.-.T., Liao, L.-.D. & Makeig, S. 2011, 'Cognition in action: imaging brain/body dynamics in mobile humans.', Reviews in the neurosciences, vol. 22, no. 6, pp. 593-608.
View/Download from: Publisher's site
We have recently developed a mobile brain imaging method (MoBI), that allows for simultaneous recording of brain and body dynamics of humans actively behaving in and interacting with their environment. A mobile imaging approach was needed to study cognitive processes that are inherently based on the use of human physical structure to obtain behavioral goals. This review gives examples of the tight coupling between human physical structure with cognitive processing and the role of supraspinal activity during control of human stance and locomotion. Existing brain imaging methods for actively behaving participants are described and new sensor technology allowing for mobile recordings of different behavioral states in humans is introduced. Finally, we review recent work demonstrating the feasibility of a MoBI system that was developed at the Swartz Center for Computational Neuroscience at the University of California, San Diego, demonstrating the range of behavior that can be investigated with this method.
Liao, L.-.D., Wang, I.-.J., Chen, S.-.F., Chang, J.-.Y. & Lin, C.-.T. 2011, 'Design, fabrication and experimental validation of a novel dry-contact sensor for measuring electroencephalography signals without skin preparation.', Sensors (Basel, Switzerland), vol. 11, no. 6, pp. 5819-5834.
View/Download from: Publisher's site
In the present study, novel dry-contact sensors for measuring electro-encephalography (EEG) signals without any skin preparation are designed, fabricated by an injection molding manufacturing process and experimentally validated. Conventional wet electrodes are commonly used to measure EEG signals; they provide excellent EEG signals subject to proper skin preparation and conductive gel application. However, a series of skin preparation procedures for applying the wet electrodes is always required and usually creates trouble for users. To overcome these drawbacks, novel dry-contact EEG sensors were proposed for potential operation in the presence or absence of hair and without any skin preparation or conductive gel usage. The dry EEG sensors were designed to contact the scalp surface with 17 spring contact probes. Each probe was designed to include a probe head, plunger, spring, and barrel. The 17 probes were inserted into a flexible substrate using a one-time forming process via an established injection molding procedure. With these 17 spring contact probes, the flexible substrate allows for high geometric conformity between the sensor and the irregular scalp surface to maintain low skin-sensor interface impedance. Additionally, the flexible substrate also initiates a sensor buffer effect, eliminating pain when force is applied. The proposed dry EEG sensor was reliable in measuring EEG signals without any skin preparation or conductive gel usage, as compared with the conventional wet electrodes.
Lin, C.-.T., Liao, L.-.D., Liu, Y.-.H., Wang, I.-.J., Lin, B.-.S. & Chang, J.-.Y. 2011, 'Novel dry polymer foam electrodes for long-term EEG measurement.', IEEE transactions on bio-medical engineering, vol. 58, no. 5, pp. 1200-1207.
View/Download from: Publisher's site
A novel dry foam-based electrode for long-term EEG measurement was proposed in this study. In general, the conventional wet electrodes are most frequently used for EEG measurement. However, they require skin preparation and conduction gels to reduce the skin-electrode contact impedance. The aforementioned procedures when wet electrodes were used usually make trouble to users easily. In order to overcome the aforesaid issues, a novel dry foam electrode, fabricated by electrically conductive polymer foam covered by a conductive fabric, was proposed. By using conductive fabric, which provides partly polarizable electric characteristic, our dry foam electrode exhibits both polarization and conductivity, and can be used to measure biopotentials without skin preparation and conduction gel. In addition, the foam substrate of our dry electrode allows a high geometric conformity between the electrode and irregular scalp surface to maintain low skin-electrode interface impedance, even under motion. The experimental results presented that the dry foam electrode performs better for long-term EEG measurement, and is practicable for daily life applications.
Li, C.H., Kuo, B.C. & Lin, C.T. 2011, 'LDA-Based clustering algorithm and its application to an unsupervised feature extraction', IEEE Transactions on Fuzzy Systems, vol. 19, no. 1, pp. 152-163.
View/Download from: Publisher's site
Research has shown fuzzy c-means (FCM) clustering to be a powerful tool to partition samples into different categories. However, the objective function of FCM is based only on the sum of distances of samples to their cluster centers, which is equal to the trace of the within-cluster scatter matrix. In this study, we propose a clustering algorithm based on both within- and between-cluster scatter matrices, extended from linear discriminant analysis (LDA), and its application to an unsupervised feature extraction (FE). Our proposed methods comprise between- and within-cluster scatter matrices modified from the between- and within-class scatter matrices of LDA. The scatter matrices of LDA are special cases of our proposed unsupervised scatter matrices. The results of experiments on both synthetic and real data show that the proposed clustering algorithm can generate similar or better clustering results than 11 popular clustering algorithms: K-means, K-medoid, FCM, the GustafsonKessel, GathGeva, possibilistic c-means (PCM), fuzzy PCM, possibilistic FCM, fuzzy compactness and separation, a fuzzy clustering algorithm based on a fuzzy treatment of finite mixtures of multivariate Students t distributions algorithms, and a fuzzy mixture of the Students t factor analyzers model. The results also show that the proposed FE outperforms principal component analysis and independent component analysis. &copy; 2006 IEEE.
Lin, C.-.T., Chen, S.-.A., Chiu, T.-.T., Lin, H.-.Z. & Ko, L.-.W. 2011, 'Spatial and temporal EEG dynamics of dual-task driving performance.', Journal of neuroengineering and rehabilitation, vol. 8, p. 11.
View/Download from: Publisher's site
BACKGROUND: Driver distraction is a significant cause of traffic accidents. The aim of this study is to investigate Electroencephalography (EEG) dynamics in relation to distraction during driving. To study human cognition under a specific driving task, simulated real driving using virtual reality (VR)-based simulation and designed dual-task events are built, which include unexpected car deviations and mathematics questions. METHODS: We designed five cases with different stimulus onset asynchrony (SOA) to investigate the distraction effects between the deviations and equations. The EEG channel signals are first converted into separated brain sources by independent component analysis (ICA). Then, event-related spectral perturbation (ERSP) changes of the EEG power spectrum are used to evaluate brain dynamics in time-frequency domains. RESULTS: Power increases in the theta and beta bands are observed in relation with distraction effects in the frontal cortex. In the motor area, alpha and beta power suppressions are also observed. All of the above results are consistently observed across 15 subjects. Additionally, further analysis demonstrates that response time and multiple cortical EEG power both changed significantly with different SOA. CONCLUSIONS: This study suggests that theta power increases in the frontal area is related to driver distraction and represents the strength of distraction in real-life situations.
Chen, P.Y., Van, L.D., Khoo, I.H., Reddy, H.C. & Lin, C.T. 2011, 'Power-efficient and cost-effective 2-D symmetry filter architectures', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 58, no. 1, pp. 112-125.
View/Download from: Publisher's site
This paper presents two-dimensional (2-D) VLSI digital filter structures possessing various symmetries in the filter magnitude response. For this purpose, four Type-1 and four Type-2 power-efficient and cost-effective 2-D magnitude symmetry filter architectures possessing diagonal, fourfold rotational, quadrantal, and octagonal symmetries with reduced number of multipliers and one power-efficient and cost-effective multimode 2-D symmetry filter are given. By combining the identities of the four Type-1 symmetry filter structures, the proposed multimode 2-D symmetry filter is capable of providing four different operation modes: diagonal symmetry mode (DSM), fourfold rotational symmetry mode (FRSM), quadrantal symmetry mode (QSM), and octagonal symmetry mode (OSM). The proposed diagonal, fourfold rotational, quadrantal, and octagonal symmetry filter structures can attain power savings of 16.77%, 36.30%, 22.90%, and 37.73% with respect to that of the conventional 2-D filter design without symmetry. On the other hand, the proposed DSM, FRSM, QSM, and OSM modes can reduce power consumption by 11.01%, 31.42%, 17.53%, and 35.26% compared with that of the conventional 2-D filter design. The proposed multimode filter can result in a 63.25% area reduction compared with the sum of the areas of the four individual Type-1 symmetry filter structures. &copy; 2010 IEEE.
Su, M.T., Chen, C.H., Lin, C.J. & Lin, C.T. 2011, 'A rule-based symbiotic modified differential evolution for self-organizing neuro-fuzzy systems', Applied Soft Computing Journal, vol. 11, no. 8, pp. 4847-4858.
View/Download from: Publisher's site
This study proposes a Rule-Based Symbiotic MOdified Differential Evolution (RSMODE) for Self-Organizing Neuro-Fuzzy Systems (SONFS). The RSMODE adopts a multi-subpopulation scheme that uses each individual represents a single fuzzy rule and each individual in each subpopulation evolves separately. The proposed RSMODE learning algorithm consists of structure learning and parameter learning for the SONFS model. The structure learning can determine whether or not to generate a new rule-based subpopulation which satisfies the fuzzy partition of input variables using the entropy measure. The parameter learning combines two strategies including a subpopulation symbiotic evolution and a modified differential evolution. The RSMODE can automatically generate initial subpopulation and each individual in each subpopulation evolves separately using a modified differential evolution. Finally, the proposed method is applied in various simulations. Results of this study demonstrate the effectiveness of the proposed RSMODE learning algorithm. &copy; 2011 Elsevier B.V. All rights reserved.
Chiu, T.-.T., Lin, C.-.L., Young, K.-.Y., Lin, C.-.T., Hsu, S.-.H., Yang, B.-.S. & Huang, Z.-.R. 2011, 'A study of Fitts' law on goal-directed aiming task with moving targets.', Perceptual and motor skills, vol. 113, no. 1, pp. 339-352.
View/Download from: Publisher's site
Most research based on Fitts' law define a log-linear relationship between temporal and spatial accuracy in goal-directed aiming tasks using stationary targets. Whether this relationship holds or not when the targets have varying velocities, and how the behavioral strategies and physical activities may change accordingly are of interest. The aim of this study was to investigate the relationship between temporal and spatial accuracy in goal-directed aiming tasks with moving targets. Participants were asked to aim at two target widths using a joystick. Results demonstrated that in a goal-directed aiming task there was a negative effect on performance when target velocity was increased or target width was decreased. Participants moved faster and then made more systematic errors in a high-velocity target condition. Results may be applicable to the complex perceptual-motor behavior of people who perform tasks using computers.
Li, C.H., Kuo, B.C., Lin, C.T. & Huang, C.S. 2011, 'A Spatial-Contextual Support Vector Machine for Remotely Sensed Image Classification', IEEE Transactions on Geoscience and Remote Sensing.
View/Download from: Publisher's site
Recent studies show that hyperspectral image classification techniques that use both spectral and spatial information are more suitable, effective, and robust than those that use only spectral information. Using a spatial-contextual term, this study modifies the decision function and constraints of a support vector machine (SVM) and proposes two kinds of spatial-contextual SVMs for hyperspectral image classification. One machine, which is based on the concept of Markov random fields (MRFs), uses the spatial information in the original space (SCSVM). The other machine uses the spatial information in the feature space (SCSVMF), i.e., the nearest neighbors in the feature space. The SCSVM is better able to classify pixels of different class labels with similar spectral values and deal with data that have no clear numerical interpretation. To evaluate the effectiveness of SCSVM, the experiments in this study compare the performances of other classifiers: an SVM, a context-sensitive semisupervised SVM, a maximum likelihood (ML) classifier, a Bayesian contextual classifier based on MRFs (ML_MRF), and k nearest neighbor classifier. Experimental results show that the proposed method achieves good classification performance on famous hyperspectral images (the Indian Pine site (IPS) and the Washington, DC mall data sets). The overall classification accuracy of the hyperspectral image of the IPS data set with 16 classes is 95.5%. The kappa accuracy is up to 94.9%, and the average accuracy of each class is up to 94.2%.
Lin, C.T., Lin, F.C., Chen, S.A., Lu, S.W., Chen, T.C. & Ko, L.W. 2010, 'EEG-based brain-computer interface for smart living environmental auto-adjustment', Journal of Medical and Biological Engineering, vol. 30, no. 4, pp. 237-245.
An EEG-based smart living environmental control system to auto-adjust the living environment is proposed in this study. Many environmental control systems have been proposed to improve human life quality in recent years. However, there is little research focused on environment control by using a human's physiological state directly. Even though some studies have proposed brain computer interface-based (BCI-based) environmental control systems, most of them encountered signal quality decline during long-term physiological monitoring with conventional wet electrodes. Moreover, such BCI-based environmental control systems are actively controlled by users; less close-loop feedback capability can be provided between environment and user for automation. Based on the advance of our technique for BCI and the improvement of micro-electro- mechanical-system-based (MEMS-based) dry electrode sensors, we combined these techniques to demonstrate an auto-adjustable living environment control system, e.g., illumination of light and fan speed of air conditioner, depends on the physiological change of the user, even for long-term physiological monitoring. The system is structured with five units: a wireless portable EEG acquisition circuit unit, an interactive flow control unit with a real-time physiology signal processing unit, which is implemented on a dual-core processor, an environment controller unit and a host system for data storage and display. The proposed system has been verified in a simulated environment and the experimental results show that the air conditioner and the lights can be successfully and automatically adjusted in real-time based on the subject's physiological changes, which indicate the proposed system can be implemented and constructed in the practical smart living environment or for other applications.
Lin, C.-.T., Chang, K.-.C., Lin, C.-.L., Chiang, C.-.C., Lu, S.-.W., Chang, S.-.S., Lin, B.-.S., Liang, H.-.Y., Chen, R.-.J., Lee, Y.-.T. & Ko, L.-.W. 2010, 'An intelligent telecardiology system using a wearable and wireless ECG to detect atrial fibrillation.', IEEE transactions on information technology in biomedicine : a publication of the IEEE Engineering in Medicine and Biology Society, vol. 14, no. 3, pp. 726-733.
View/Download from: Publisher's site
This study presents a novel wireless, ambulatory, real-time, and autoalarm intelligent telecardiology system to improve healthcare for cardiovascular disease, which is one of the most prevalent and costly health problems in the world. This system consists of a lightweight and power-saving wireless ECG device equipped with a built-in automatic warning expert system. This device is connected to a mobile and ubiquitous real-time display platform. The acquired ECG signals are instantaneously transmitted to mobile devices, such as netbooks or mobile phones through Bluetooth, and then, processed by the expert system. An alert signal is sent to the remote database server, which can be accessed by an Internet browser, once an abnormal ECG is detected. The current version of the expert system can identify five types of abnormal cardiac rhythms in real-time, including sinus tachycardia, sinus bradycardia, wide QRS complex, atrial fibrillation (AF), and cardiac asystole, which is very important for both the subjects who are being monitored and the healthcare personnel tracking cardiac-rhythm disorders. The proposed system also activates an emergency medical alarm system when problems occur. Clinical testing reveals that the proposed system is approximately 94% accurate, with high sensitivity, specificity, and positive prediction rates for ten normal subjects and 20 AF patients. We believe that in the future a business-card-like ECG device, accompanied with a mobile phone, can make universal cardiac protection service possible.
Shou, Y.W., Lin, C.T., Siana, L. & Yang, C.T. 2010, 'Multiclient identification system using adaptive probabilistic model', Eurasip Journal on Advances in Signal Processing, vol. 2010.
View/Download from: Publisher's site
This paper aims at integrating detection and identification of human faces in a more practical and real-time face recognition system. The proposed face detection system is based on the cascade Adaboost method to improve the precision and robustness toward unstable surrounding lightings. Our Adaboost method innovates to adjust the environmental lighting conditions by histogram lighting normalization and to accurately locate the face regions by a region-based-clustering process as well. We also address on the problem of multi-scale faces in this paper by using 12 different scales of searching windows and 5 different orientations for each client in pursuit of the multi-view independent face identification. There are majorly two methodological parts in our face identification system, including PCA (principal component analysis) facial feature extraction and adaptive probabilistic model (APM). The structure of our implemented APM with a weighted combination of simple probabilistic functions constructs the likelihood functions by the probabilistic constraint in the similarity measures. In addition, our proposed method can online add a new client and update the information of registered clients due to the constructed APM. The experimental results eventually show the superior performance of our proposed system for both offline and real-time online testing. Copyright &copy; 2010 Chin-Teng Lin et al.
Chang, C.W., Ko, L.W., Lin, F.C., Su, T.P., Jung, T.P., Lin, C.T. & Chiou, J.C. 2010, 'Drowsiness monitoring with EEG-based MEMS biosensing technologies', GeroPsych: The Journal of Gerontopsychology and Geriatric Psychiatry, vol. 23, no. 2, pp. 107-113.
View/Download from: Publisher's site
Electroencephalography (EEG) has been widely adopted to monitor changes in cognitive states, particularly stages of sleep, as EEG recordings contain a wealth of information reflecting changes in alertness and sleepiness. In this study, silicon dry electrodes based on Micro-Electro-Mechanical Systems (MEMS) were developed to bring high-quality EEG acquisition to operational workplaces. They have superior conductivity performance, large signal intensity, and are smaller in size than conventional (wet) electrodes. An EEG-based drowsiness estimation system consisting of a dry-electrode array, power spectrum estimation, principal component analysis (PCA)-based EEG signal analysis, and multivariate linear regression was developed to estimate drivers' drowsiness levels in a virtual-reality-based dynamic driving simulator. The proposed system can help elders who are often affected by periods of tiredness and fatigue. &copy; 2010 Hogrefe Publishing.
Shou, Y.W., Lin, C.T. & Shen, T.K. 2010, 'Construction of fisheye lens inverse perspective mapping model and its applications of obstacle detection', Eurasip Journal on Advances in Signal Processing, vol. 2010.
View/Download from: Publisher's site
In this paper, we develop a vision based obstacle detection system by utilizing our proposed fisheye lens inverse perspective mapping (FLIPM) method. The new mapping equations are derived to transform the images captured by the fisheye lens camera into the undistorted remapped ones under practical circumstances. In the obstacle detection, we make use of the features of vertical edges on objects from remapped images to indicate the relative positions of obstacles. The static information of remapped images in the current frame is referred to determining the features of source images in the searching stage from either the profile or temporal IPM difference image. The profile image can be acquired by several processes such as sharpening, edge detection, morphological operation, and modified thinning algorithms on the remapped image. The temporal IPM difference image can be obtained by a spatial shift on the remapped image in the previous frame. Moreover, the polar histogram and its post-processing procedures will be used to indicate the position and length of feature vectors and to remove noises as well. Our obstacle detection can give drivers the warning signals within a limited distance from nearby vehicles while the detected obstacles are even with the quasi-vertical edges. Copyright &copy; 2010 Chin-Teng Lin et al.
Liao, L.-.D., Li, M.-.L., Lai, H.-.Y., Shih, Y.-.Y.I., Lo, Y.-.C., Tsang, S., Chao, P.C.-.P., Lin, C.-.T., Jaw, F.-.S. & Chen, Y.-.Y. 2010, 'Imaging brain hemodynamic changes during rat forepaw electrical stimulation using functional photoacoustic microscopy.', NeuroImage, vol. 52, no. 2, pp. 562-570.
View/Download from: Publisher's site
The present study reported the development of a novel functional photoacoustic microscopy (fPAM) system for investigating hemodynamic changes in rat cortical vessels associated with electrical forepaw stimulation. Imaging of blood optical absorption by fPAM at multiple appropriately-selected and distinct wavelengths can be used to probe changes in total hemoglobin concentration (HbT, i.e., cerebral blood volume [CBV]) and hemoglobin oxygen saturation (SO(2)). Changes in CBV were measured by images acquired at a wavelength of 570nm (lambda(570)), an isosbestic point of the molar extinction spectra of oxy- and deoxy-hemoglobin, whereas SO(2) changes were sensed by pixel-wise normalization of images acquired at lambda(560) or lambda(600) to those at lambda(570). We demonstrated the capacity of the fPAM system to image and quantify significant contralateral changes in both SO(2) and CBV driven by electrical forepaw stimulation. The fPAM system complements existing imaging techniques, with the potential to serve as a favorable tool for explicitly studying brain hemodynamics in animal models.
Lin, C.T., Chang, C.J., Lin, B.S., Hung, S.H., Chao, C.F. & Wang, I.J. 2010, 'A real-time wireless brain-computer interface system for drowsiness detection', IEEE Transactions on Biomedical Circuits and Systems, vol. 4, no. 4, pp. 214-222.
View/Download from: Publisher's site
A real-time wireless electroencephalogram (EEG)-based braincomputer interface (BCI) system for drowsiness detection has been proposed. Drowsy driving has been implicated as a causal factor in many accidents. Therefore, real-time drowsiness monitoring can prevent traffic accidents effectively. However, current BCI systems are usually large and have to transmit an EEG signal to a back-end personal computer to process the EEG signal. In this study, a novel BCI system was developed to monitor the human cognitive state and provide biofeedback to the driver when drowsy state occurs. The proposed system consists of a wireless physiological signal-acquisition module and an embedded signal-processing module. Here, the physiological signal-acquisition module and embedded signal-processing module were designed for long-term EEG monitoring and real-time drowsiness detection, respectively. The advantages of low ower consumption and small volume of the proposed system are suitable for car applications. Moreover, a real-time drowsiness detection algorithm was also developed and implemented in this system. The experiment results demonstrated the feasibility of our proposed BCI system in a practical driving application. &copy; 2010 IEEE.
Shou, Y.W., Lin, C.T., Yang, C.T. & Shen, T.K. 2010, 'An efficient and robust moving shadow removal algorithm and its applications in ITS', Eurasip Journal on Advances in Signal Processing, vol. 2010.
View/Download from: Publisher's site
We propose an efficient algorithm for removing shadows of moving vehicles caused by non-uniform distributions of light reflections in the daytime. This paper presents a brand-new and complete structure in feature combination as well as analysis for orientating and labeling moving shadows so as to extract the defined objects in foregrounds more easily in each snapshot of the original files of videos which are acquired in the real traffic situations. Moreover, we make use of Gaussian Mixture Model (GMM) for background removal and detection of moving shadows in our tested images, and define two indices for characterizing non-shadowed regions where one indicates the characteristics of lines and the other index can be characterized by the information in gray scales of images which helps us to build a newly defined set of darkening ratios (modified darkening factors) based on Gaussian models. To prove the effectiveness of our moving shadow algorithm, we carry it out with a practical application of traffic flow detection in ITS (Intelligent Transportation System)-vehicle counting. Our algorithm shows the faster processing speed, 13.84ms/frame, and can improve the accuracy rate in 4% 10% for our three tested videos in the experimental results of vehicle counting. Copyright &copy; 2010 Chin-Teng Lin et al.
Lin, C.-.T., Huang, K.-.C., Chao, C.-.F., Chen, J.-.A., Chiu, T.-.W., Ko, L.-.W. & Jung, T.-.P. 2010, 'Tonic and phasic EEG and behavioral changes induced by arousing feedback.', NeuroImage, vol. 52, no. 2, pp. 633-642.
View/Download from: Publisher's site
This study investigates brain dynamics and behavioral changes in response to arousing auditory signals presented to individuals experiencing momentary cognitive lapses during a sustained-attention task. Electroencephalographic (EEG) and behavioral data were simultaneously collected during virtual-reality (VR) based driving experiments, in which subjects were instructed to maintain their cruising position and compensate for randomly induced lane deviations using the steering wheel. 30-channel EEG data were analyzed by independent component analysis and the short-time Fourier transform. Across subjects and sessions, intermittent performance during drowsiness was accompanied by characteristic spectral augmentation or suppression in the alpha- and theta-band spectra of a bilateral occipital component, corresponding to brief periods of normal (wakeful) and hypnagogic (sleeping) awareness and behavior. Arousing auditory feedback was delivered to the subjects in half of the non-responded lane-deviation events, which immediately agitated subject's responses to the events. The improved behavioral performance was accompanied by concurrent spectral suppression in the theta- and alpha-bands of the bilateral occipital component. The effects of auditory feedback on spectral changes lasted 30s or longer. The results of this study demonstrate the amount of cognitive state information that can be extracted from noninvasively recorded EEG data and the feasibility of online assessment and rectification of brain networks exhibiting characteristic dynamic patterns in response to momentary cognitive challenges.
Tsai, C.H., Liao, L.D., Luo, Y.S., Chao, P.C.P., Chen, E.C., Meng, H.F., Chen, W.D., Lin, S.K. & Lin, C.T. 2010, 'Optimal design and fabrication of ITO/organic photonic crystals in polymer light-emitting diodes using a focused ion beam', Microelectronic Engineering, vol. 87, no. 5-8, pp. 1331-1335.
View/Download from: Publisher's site
This study aims to achieve high directionality and large extraction of light emission from polymer light-emitting diodes via optimizing photonic crystals (PCs). The optimization is achieved through the optical modeling using a 3D finite-difference time-domain (FDTD) method and the intelligent numerical optimization technique, genetic algorithm (GA). In results, the PC period, depth and filling factor can be found. The FDTD method has been proven very effective for modeling complex multilayer PLED devices in which the electron-transporting layer is only a few tens of nanometers away from a metallic layer. On the other hand, GA is a powerful tool to cope with a complicated optimization problem with multiple variables to optimize. The optimized PCs are later fabricated by focused ion beams (FIB) for the PLED. With desired PCs, an increase of the extraction efficiency in excess of 46% is achieved experimentally and the 3D FDTD calculation explains this result faithfully. &copy; 2009 Elsevier B.V. All rights reserved.
Lin, K.-.L., Lin, C.-.T. & Pal, N.R. 2010, 'Incremental Mountain Clustering Method to find building blocks for constructing structures of proteins.', IEEE transactions on nanobioscience, vol. 9, no. 4, pp. 278-288.
View/Download from: Publisher's site
In this paper we propose an algorithm named Incremental Structural Mountain Clustering Method (ISMCM) with a view to finding a library of building blocks for reconstruction of 3-D structures of proteins/peptides. The building blocks are short structural motifs that are identified based on an estimate of local "density" of 3-D fragments computed using a measure of structural similarity. The structural similarity is computed after the best-molecular-fit alignment of pairs of fragments. The algorithm is tested on two well known benchmark data sets. Following the protocols used by other researchers, for the first data set we reconstruct a set of 71 test peptides (up to first 60 residues) whereas for the second data set we reconstruct all 143 test peptides. The ISMCM algorithm is found to successfully reconstruct the test peptides in terms of both global-fit root-mean-square (RMS) error and local-fit RMS error. The low values of local-fit RMS errors suggest that these building blocks extracted by ISMCM are good quantizers, which can represent nearby fragments quite accurately. To further assess the quality of building blocks we use two alternative graphical ways. We also use Shannon's entropy to show the structural similarity of the clusters found by our algorithm. This is important as building blocks that represent clusters with structurally similar fragments will be very effective in reconstruction. The entropic analysis reveals a very interesting fact that the secondary structure of the central residue of the fragments in a cluster is most strongly conserved (minimum entropy) over the cluster, which might be an indicator that central residue of the structural motif plays a dominant role in local folding.
Lin, C.T. & Huang, Y.M. 2010, 'IEEE CIS DLP tour in Taiwan', IEEE Computational Intelligence Magazine, vol. 5, no. 4.
View/Download from: Publisher's site
Taiwan hosted an international workshop from January 31 to February 6, 2010, to promote the technical activities of IEEE CIS DLP (IEEE Computational Intelligence Society Distinguished Lecturer Program). This included technical talks on 'Anomaly Detection in Wireless Sensor Networks with Visual Clustering' by Jim Bezdek and 'Recognition Technology in Eldercare' by Jim Keller. The first presentation on Fuzzy Systems was held at National Chiao Tung University in Hsinchu. The second stop was the Brain Research Center at National Chiao-Tung University where Jim Bezdek and Jim Keller visited Chin-Teng Lin's laboratory and conversed with his students. Next, Jim Bezdek and Jim Keller visited the Eco-City at National Chiao-Tung University. The fourth location was National Cheng Kung University in Tainan.
Lin, C.-.T., Ko, L.-.W., Chang, M.-.H., Duann, J.-.R., Chen, J.-.Y., Su, T.-.P. & Jung, T.-.P. 2010, 'Review of wireless and wearable electroencephalogram systems and brain-computer interfaces--a mini-review.', Gerontology, vol. 56, no. 1, pp. 112-119.
View/Download from: Publisher's site
Biomedical signal monitoring systems have rapidly advanced in recent years, propelled by significant advances in electronic and information technologies. Brain-computer interface (BCI) is one of the important research branches and has become a hot topic in the study of neural engineering, rehabilitation, and brain science. Traditionally, most BCI systems use bulky, wired laboratory-oriented sensing equipments to measure brain activity under well-controlled conditions within a confined space. Using bulky sensing equipments not only is uncomfortable and inconvenient for users, but also impedes their ability to perform routine tasks in daily operational environments. Furthermore, owing to large data volumes, signal processing of BCI systems is often performed off-line using high-end personal computers, hindering the applications of BCI in real-world environments. To be practical for routine use by unconstrained, freely-moving users, BCI systems must be noninvasive, nonintrusive, lightweight and capable of online signal processing. This work reviews recent online BCI systems, focusing especially on wearable, wireless and real-time systems.
Chen, Y.-.C., Duann, J.-.R., Chuang, S.-.W., Lin, C.-.L., Ko, L.-.W., Jung, T.-.P. & Lin, C.-.T. 2010, 'Spatial and temporal EEG dynamics of motion sickness.', NeuroImage, vol. 49, no. 3, pp. 2862-2870.
View/Download from: Publisher's site
This study investigates motion-sickness-related brain responses using a VR-based driving simulator on a motion platform with six degrees of freedom, which provides both visual and vestibular stimulations to induce motion sickness in a manner that is close to that in daily life. Subjects' brain dynamics associated with motion sickness were measured using a 32-channel EEG system. Their degree of motion sickness was simultaneously and continuously reported using an onsite joystick, providing non-stop behavioral references to the recorded EEG changes. The acquired EEG signals were parsed by independent component analysis (ICA) into maximally independent processes. The decomposition enables the brain dynamics that are induced by the motion of the platform and motion sickness to be disassociated. Five MS-related brain processes with equivalent dipoles located in the left motor, the parietal, the right motor, the occipital and the occipital midline areas were consistently identified across all subjects. The parietal and motor components exhibited significant alpha power suppression in response to vestibular stimuli, while the occipital components exhibited MS-related power augmentation in mainly theta and delta bands; the occipital midline components exhibited a broadband power increase. Further, time series cross-correlation analysis was employed to evaluate relationships between the spectral changes associated with different brain processes and the degree of motion sickness. According to our results, it is suggested both visual and vestibular stimulations should be used to induce motion sickness in brain dynamic studies.
Shou, Y.W., Lin, C.T., Siana, L. & Shen, T.K. 2010, 'A conditional entropy-based independent component analysis for applications in human detection and tracking', Eurasip Journal on Advances in Signal Processing, vol. 2010.
View/Download from: Publisher's site
We present in this paper a modified independent component analysis (mICA) based on the conditional entropy to discriminate unsorted independent components. We make use of the conditional entropy to select an appropriate subset of the ICA features with superior capability in classification and apply support vector machine (SVM) to recognizing patterns of human and nonhuman. Moreover, we use the models of background images based on Gaussian mixture model (GMM) to handle images with complicated backgrounds. Also, the color-based shadow elimination and head models in ellipse shapes are combined to improve the performance of moving objects extraction and recognition in our system. Our proposed tracking mechanism monitors the movement of humans, animals, or vehicles within a surveillance area and keeps tracking the moving pedestrians by using the color information in HSV domain. Our tracking mechanism uses the Kalman filter to predict locations of moving objects for the conditions in lack of color information of detected objects. Finally, our experimental results show that our proposed approach can perform well for real-time applications in both indoor and outdoor environments. Copyright &copy; 2010 Chin-Teng Lin et al.
Lin, K.L., Lin, C.T., Pal, N.R. & Ojha, S. 2009, 'Structural building blocks: construction of protein 3-D structures using a structural variant of mountain clustering method.', IEEE engineering in medicine and biology magazine : the quarterly magazine of the Engineering in Medicine & Biology Society, vol. 28, no. 4, pp. 38-44.
Chen, C.H., Lin, C.L. & Lin, C.T. 2009, 'Nonlinear system control using adaptive neural fuzzy networks based on a modified differential evolution', IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews, vol. 39, no. 4, pp. 459-473.
View/Download from: Publisher's site
This study presents an adaptive neural fuzzy network (ANFN) controller based on a modified differential evolution (MODE) for solving control problems. The proposed ANFN controller adopts a functional link neural network as the consequent part of the fuzzy rules. Thus, the consequent part of the ANFN controller is a nonlinear combination of input variables. The proposed MODE learning algorithm adopts an evolutionary learning method to optimize the controller parameters. For design optimization, a new criterion is introduced. A hardware-in-the loop control technique is developed and applied to the designed ANFN controller using the MODE learning algorithm. The proposed ANFN controller with the MODE learning algorithm (ANFN-MODE) is used in two practical applications-the planetary-train-type inverted pendulum system and the magnetic levitation system. The experiment is developed in a real-time visual simulation environment. Experimental results of this study have demonstrated the robustness and effectiveness of the proposed ANFN-MODE controller. &copy; 2009 IEEE.
Lin, C.T., Ko, L.W. & Shen, T.K. 2009, 'Computational intelligent brain computer interaction and its applications on driving cognition', IEEE Computational Intelligence Magazine, vol. 4, no. 4, pp. 32-46.
View/Download from: Publisher's site
Driving is one of the most common attention-demanding tasks in daily life. Driver's fatigue, drowsiness, inattention, and distraction are reported a major causal factor in many traffic accidents. Due to the drivers lost their attention, they had markedly reduced the perception, recognition and vehicle control abilities. In recent years, many computational intelligent technologies were developed for preventing traffic accidents caused by driver's inattention. Driver's drowsiness and distraction related studies had become a major interest research topic in automotive safety engineering. Many researches had investigated the driving cognition in cognitive neuro-engineering, but how to utilize the main findings of driving-related cognitive researches in traditional cognitive neuroscience and integrate with computational intelligence technologies for augmenting driving performance will become a big challenge in the interdisciplinary research area. For this reason, we attempt to integrate the driving cognition for real life application in this study. The implications of the driving cognition in cognitive neuroscience and computational intelligence for daily applications are also demonstrated through two common attention-related driving studies: (1) cognitive-state monitoring of the driver performing the realistic long-term driving tasks in a simulated realistic-driving environment; and (2) to extract the brain dynamic changes of driver's distraction effect during dual-task driving. Experimental results of these studies provide new insights into the understanding of complex brain functions of participants actively performing ordinary tasks in natural body positions and situations within real operational environments. &copy; 2006 IEEE.
Lin, C.-.T., Chiu, T.-.T., Huang, T.-.Y., Chao, C.-.F., Liang, W.-.C., Hsu, S.-.H. & Ko, L.-.W. 2009, 'Assessing effectiveness of various auditory warning signals in maintaining drivers' attention in virtual reality-based driving environments.', Perceptual and motor skills, vol. 108, no. 3, pp. 825-835.
View/Download from: Publisher's site
Drivers' fatigue contributes to traffic accidents, so drivers must maintain adequate alertness. The effectiveness of audio alarms in maintaining driving performance and characteristics of alarms was studied in a virtural reality-based driving environment. Response time to the car's drifting was measured under seven conditions: with no warnings and with continuous warning tones (500 Hz, 1750 Hz, and 3000 Hz), and with tone bursts at 500 Hz, 1750 Hz, and 3000 Hz. Analyses showed the audio warning signals significantly improved driving. Further, the tones' spectral characteristics significantly influenced the effectiveness of the warning.
Lin, C.J., Chen, C.H. & Lin, C.T. 2009, 'A hybrid of cooperative particle swarm optimization and cultural algorithm for neural fuzzy networks and its prediction applications', IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews, vol. 39, no. 1, pp. 55-68.
View/Download from: Publisher's site
This study presents an evolutionary neural fuzzy network, designed using the functional-link-based neural fuzzy network (FLNFN) and a new evolutionary learning algorithm. This new evolutionary learning algorithm is based on a hybrid of cooperative particle swarm optimization and cultural algorithm. It is thus called cultural cooperative particle swarm optimization (CCPSO). The proposed CCPSO method, which uses cooperative behavior among multiple swarms, can increase the global search capacity using the belief space. Cooperative behavior involves a collection of multiple swarms that interact by exchanging information to solve a problem. The belief space is the information repository in which the individuals can store their experiences such that other individuals can learn from them indirectly. The proposed FLNFN model uses functional link neural networks as the consequent part of the fuzzy rules. This study uses orthogonal polynomials and linearly independent functions in a functional expansion of the functional link neural networks. The FLNFN model can generate the consequent part of a nonlinear combination of input variables. Finally, the proposed FLNFN with CCPSO (FLNFN-CCPSO) is adopted in several predictive applications. Experimental results have demonstrated that the proposed CCPSO method performs well in predicting the time series problems. &copy; 2008 IEEE.
Lin, C.T., Hong, C.T. & Yang, C.T. 2009, 'Real-time digital image stabilization system using modified proportional integrated controller', IEEE Transactions on Circuits and Systems for Video Technology, vol. 19, no. 3, pp. 427-431.
View/Download from: Publisher's site
This paper presents a novel, real-time digital image stabilization (DIS) system using a modified proportional integrated (MPI) controller to remove stably unwanted shaking from an image sequence that is captured by a hand-held video camera without affecting the deliberate panning motion of the camera. Researchers have addressed the trade-off problem between the removal of shaking and the preservation of global camera movement. This paper analyzes motion compensation, which is formulated as a control problem. A compensating motion vector (CMV) estimation method with a modified PI control system is proposed both to remove the unwanted jitter and to preserve the deliberate, panning motion of the camera. Experimental results demonstrate that the proposed system provides robust motion compensation of image sequences in various conditions. &copy; 2006 IEEE.
Lin, C.-.T., Chen, Y.-.C., Huang, T.-.Y., Chiu, T.-.T., Ko, L.-.W., Liang, S.-.F., Hsieh, H.-.Y., Hsu, S.-.H. & Duann, J.-.R. 2008, 'Development of wireless brain computer interface with embedded multitask scheduling and its application on real-time driver's drowsiness detection and warning.', IEEE transactions on bio-medical engineering, vol. 55, no. 5, pp. 1582-1591.
View/Download from: Publisher's site
Biomedical signal monitoring systems have been rapidly advanced with electronic and information technologies in recent years. However, most of the existing physiological signal monitoring systems can only record the signals without the capability of automatic analysis. In this paper, we proposed a novel brain-computer interface (BCI) system that can acquire and analyze electroencephalogram (EEG) signals in real-time to monitor human physiological as well as cognitive states, and, in turn, provide warning signals to the users when needed. The BCI system consists of a four-channel biosignal acquisition/amplification module, a wireless transmission module, a dual-core signal processing unit, and a host system for display and storage. The embedded dual-core processing system with multitask scheduling capability was proposed to acquire and process the input EEG signals in real time. In addition, the wireless transmission module, which eliminates the inconvenience of wiring, can be switched between radio frequency (RF) and Bluetooth according to the transmission distance. Finally, the real-time EEG-based drowsiness monitoring and warning algorithms were implemented and integrated into the system to close the loop of the BCI system. The practical online testing demonstrates the feasibility of using the proposed system with the ability of real-time processing, automatic analysis, and online warning feedback in real-world operation and living environments.
Chung, I.F., Lin, C.T., Lin, K.L., Ko, L.W., Liang, S.F. & Kuo, B.C. 2008, 'Nonparametric single-trial EEG feature extraction and classification of driver's cognitive responses', Eurasip Journal on Advances in Signal Processing, vol. 2008.
View/Download from: Publisher's site
We proposed an electroencephalographic (EEG) signal analysis approach to investigate the driver's cognitive response to traffic-light experiments in a virtual-reality-(VR-) based simulated driving environment. EEG signals are digitally sampled and then transformed by three different feature extraction methods including nonparametric weighted feature extraction (NWFE), principal component analysis (PCA), and linear discriminant analysis (LDA), which were also used to reduce the feature dimension and project the measured EEG signals to a feature space spanned by their eigenvectors. After that, the mapped data could be classified with fewer features and their classification results were compared by utilizing two different classifiers including k nearest neighbor classification (KNNC) and naive bayes classifier (NBC). Experimental data were collected from 6 subjects and the results show that NWFE+NBC gives the best classification accuracy ranging from 71%77%, which is over 10%24% higher than LDA+KNN1. It also demonstrates the feasibility of detecting and analyzing single-trial EEG signals that represent operators' cognitive states and responses to task events.
Lin, C.T. & Siana, L. 2008, 'An efficient human detection system using adaptive neural fuzzy networks', International Journal of Fuzzy Systems, vol. 10, no. 3, pp. 150-160.
The proposed efficient human detection system is based on an adaptive neural fuzzy network (ANFN). In the preprocessing process, we apply a background subtraction algorithm with Gaussian mixture model (GMM) background model to extract moving objects, and adopt a shadow elimination process to eliminate some noise and irregular moving objects. The modified independent component analysis (mICA) based conditional entropy is presented to extract and select the efficient features (independent components). Furthermore, we use an adaptive neural fuzzy network as a human detection system to recognize human objects. The ANFN model uses a functional link neural network (FLNN) to create the consequent part of the fuzzy rules. The orthogonal polynomial is applied as a functional expansion of the FLNN. The learning process of ANFN consists of structure learning and parameter learning. The structure learning depends on the entropy measure to determine the number of fuzzy rules. The parameter learning, based on the back propagation method, is used to adjust the membership function and corresponding weights of the FLNN. Finally, the proposed human detection system is applied in various circumstances. The results of this study demonstrate the accuracy of the proposed method. &copy; 2008 TFSA.
Chen, C.H., Lin, C.J. & Lin, C.T. 2008, 'An efficient quantum neuro-fuzzy classifier based on fuzzy entropy and compensatory operation', Soft Computing, vol. 12, no. 6, pp. 567-583.
View/Download from: Publisher's site
In this paper, a quantum neuro-fuzzy classifier (QNFC) for classification applications is proposed. The proposed QNFC model is a five-layer structure, which combines the compensatory-based fuzzy reasoning method with the traditional Takagi-Sugeno-Kang (TSK) fuzzy model. The compensatory-based fuzzy reasoning method uses adaptive fuzzy operations of neuro-fuzzy systems that can make the fuzzy logic system more adaptive and effective. Layer 2 of the QNFC model contains quantum membership functions, which are multilevel activation functions. Each quantum membership function is composed of the sum of sigmoid functions shifted by quantum intervals. A self-constructing learning algorithm, which consists of the self-clustering algorithm (SCA), quantum fuzzy entropy and the backpropagation algorithm, is also proposed. The proposed SCA method is a fast, one-pass algorithm that dynamically estimates the number of clusters in an input data space. Quantum fuzzy entropy is employed to evaluate the information on pattern distribution in the pattern space. With this information, we can determine the number of quantum levels. The backpropagation algorithm is used to tune the adjustable parameters. The simulation results have shown that (1) the QNFC model converges quickly; (2) the QNFC model has a higher correct classification rate than other models. &copy; Springer-Verlag 2007.
Liang, S.F., Lu, S.M., Chang, J.Y. & Lin, C.T. 2008, 'A novel two-stage impulse noise removal technique based on neural networks and fuzzy decision', IEEE Transactions on Fuzzy Systems, vol. 16, no. 4, pp. 863-873.
View/Download from: Publisher's site
In this paper, a novel two-stage noise removal algorithm to deal with impulse noise is proposed. In the first stage, an adaptive two-level feedforward neural network (NN) with a backpropagation training algorithm was applied to remove the noise cleanly and keep the uncorrupted information well. In the second stage, the fuzzy decision rules inspired by the human visual system (HVS) are proposed to classify the image pixels into human perception sensitive class and nonsensitive class, and to compensate the blur of the edge and the destruction caused by the median filter. An NN is proposed to enhance the sensitive regions with higher visual quality. According to the experimental results, the proposed method is superior to conventional methods in perceptual image quality as well as the clarity and smoothness in edge regions. &copy; 2008 IEEE.
Lin, C.T., Pal, N.R., Chuang, C.Y., Ko, L.W., Chao, C.F., Jung, T.P. & Liang, S.F. 2008, 'EEG-based subject- and session-independent drowsiness detection: An unsupervised approach', Eurasip Journal on Advances in Signal Processing, vol. 2008.
View/Download from: Publisher's site
Monitoring and prediction of changes in the human cognitive states, such as alertness and drowsiness, using physiological signals are very important for drivers safety. Typically, physiological studies on real-time detection of drowsiness usually use the same model for all subjects. However, the relatively large individual variability in EEG dynamics relating to loss of alertness implies that for many subjects, group statistics may not be useful to accurately predict changes in cognitive states. Researchers have attempted to build subject-dependent models based on his/her pilot data to account for individual variability. Such approaches cannot account for the cross-session variability in EEG dynamics, which may cause problems due to various reasons including electrode displacements, environmental noises, and skin-electrode impedance. Hence, we propose an unsupervised subject- and session-independent approach for detection departure from alertness in this study. Experimental results showed that the EEG power in the alpha-band (as well as in the theta-band) is highly correlated with changes in the subjects cognitive state with respect to drowsiness as reflected through his driving performance. This approach being an unsupervised and session-independent one could be used to develop a useful system for noninvasive monitoring of the cognitive state of human operators in attention-critical settings.
Tsai, Y.-.S., Chung, I.-.F., Simpson, J.C., Lee, M.-.I., Hsiung, C.-.C., Chiu, T.-.Y., Kao, L.-.S., Chiu, T.-.C., Lin, C.-.T., Lin, W.-.C., Liang, S.-.F. & Lin, C.-.C. 2008, 'Automated recognition system to classify subcellular protein localizations in images of different cell lines acquired by different imaging systems.', Microscopy research and technique, vol. 71, no. 4, pp. 305-314.
View/Download from: Publisher's site
Systemic analysis of subcellular protein localization (location proteomics) provides clues for understanding gene functions and physiological condition of the cells. However, recognition of cell images of subcellular structures highly depends on experience and becomes the rate-limiting step when classifying subcellular protein localization. Several research groups have extracted specific numerical features for the recognition of subcellular protein localization, but these recognition systems are restricted to images of single particular cell line acquired by one specific imaging system and not applied to recognize a range of cell image sources. In this study, we establish a single system for automated subcellular structure recognition to identify cell images from various sources. Two different sources of cell images, 317 Vero (http://gfp-cdna.embl.de) and 875 CHO cell images of subcellular structures, were used to train and test the system. When the system was trained by a single source of images, the recognition rate is high and specific to the trained source. The system trained by the CHO cell images gave high average recognition accuracy for CHO cells of 96%, but this was reduced to 46% with Vero images. When we trained the system using a mixture of CHO and Vero cell images, an average accuracy of recognition reached 86.6% for both CHO and Vero cell images. The system can reject images with low confidence and identify the cell images correctly recognized to avoid manual reconfirmation. In summary, we have established a single system that can recognize subcellular protein localizations from two different sources for location-proteomic studies. studies.
Tsai, Y.-.S., Lin, C.-.T., Tseng, G.C., Chung, I.-.F. & Pal, N.R. 2008, 'Discovery of dominant and dormant genes from expression data using a novel generalization of SNR for multi-class problems.', BMC bioinformatics, vol. 9, p. 425.
View/Download from: Publisher's site
BACKGROUND: The Signal-to-Noise-Ratio (SNR) is often used for identification of biomarkers for two-class problems and no formal and useful generalization of SNR is available for multiclass problems. We propose innovative generalizations of SNR for multiclass cancer discrimination through introduction of two indices, Gene Dominant Index and Gene Dormant Index (GDIs). These two indices lead to the concepts of dominant and dormant genes with biological significance. We use these indices to develop methodologies for discovery of dominant and dormant biomarkers with interesting biological significance. The dominancy and dormancy of the identified biomarkers and their excellent discriminating power are also demonstrated pictorially using the scatterplot of individual gene and 2-D Sammon's projection of the selected set of genes. Using information from the literature we have shown that the GDI based method can identify dominant and dormant genes that play significant roles in cancer biology. These biomarkers are also used to design diagnostic prediction systems. RESULTS AND DISCUSSION: To evaluate the effectiveness of the GDIs, we have used four multiclass cancer data sets (Small Round Blue Cell Tumors, Leukemia, Central Nervous System Tumors, and Lung Cancer). For each data set we demonstrate that the new indices can find biologically meaningful genes that can act as biomarkers. We then use six machine learning tools, Nearest Neighbor Classifier (NNC), Nearest Mean Classifier (NMC), Support Vector Machine (SVM) classifier with linear kernel, and SVM classifier with Gaussian kernel, where both SVMs are used in conjunction with one-vs-all (OVA) and one-vs-one (OVO) strategies. We found GDIs to be very effective in identifying biomarkers with strong class specific signatures. With all six tools and for all data sets we could achieve better or comparable prediction accuracies usually with fewer marker genes than results reported in the literature using the same computation...
Lin, C.J., Chen, C.H. & Lin, C.T. 2008, 'Efficient self-evolving evolutionary learning for neurofuzzy inference systems', IEEE Transactions on Fuzzy Systems, vol. 16, no. 6, pp. 1476-1490.
View/Download from: Publisher's site
This study proposes an efficient self-evolving evolutionary learning algorithm (SEELA) for neurofuzzy inference systems (NFISs). The major feature of the proposed SEELA is that it is based on evolutionary algorithms that can determine the number of fuzzy rules and adjust the NFIS parameters. The SEELA consists of structure learning and parameter learning. The structure learning attempts to determine the number of fuzzy rules. A subgroup symbiotic evolution is adopted to yield several variable fuzzy systems, and an elite-based structure strategy is adopted to find a suitable number of fuzzy rules for solving a problem. The parameter learning is to adjust parameters of the NFIS. It is a hybrid evolutionary algorithm of cooperative particle swarm optimization (CPSO) and cultural algorithm, called cultural CPSO (CCPSO). The CCPSO, which uses cooperative behavior among multiple swarms, can increase the global search capacity using the belief space. Experimental results demonstrate that the proposed method performs well in predicting time series and solving nonlinear control problems. &copy; 2008 IEEE.
Chen, C.H., Lin, C.J. & Lin, C.T. 2008, 'A functional-link-based neurofuzzy network for nonlinear system control', IEEE Transactions on Fuzzy Systems, vol. 16, no. 5, pp. 1362-1378.
View/Download from: Publisher's site
This study presents a functional-link-based neurofuzzy network (FLNFN) structure for nonlinear system control. The proposed FLNFN model uses a functional link neural network (FLNN) to the consequent part of the fuzzy rules. This study uses orthogonal polynomials and linearly independent functions in a functional expansion of the FLNN. Thus, the consequent part of the proposed FLNFN model is a nonlinear combination of input variables. An online learning algorithm, which consists of structure learning and parameter learning, is also presented. The structure learning depends on the entropy measure to determine the number of fuzzy rules. The parameter learning, based on the gradient descent method, can adjust the shape of the membership function and the corresponding weights of the FLNN. Furthermore, results for the universal approximator and a convergence analysis of the FLNFN model are proven. Finally, the FLNFN model is applied in various simulations. Results of this study demonstrate the effectiveness of the proposed FLNFN model. &copy; 2008 IEEE.
Lin, C.T., Yu, Y.C. & Van, L.D. 2008, 'Cost-effective triple-mode reconfigurable pipeline FFT/IFFT/2-D DCT processor', IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 16, no. 8, pp. 1058-1071.
View/Download from: Publisher's site
This investigation proposes a novel radix-42algorithm with the low computational complexity of a radix-16 algorithm but the lower hardware requirement of a radix-4 algorithm. The proposed pipeline radix-42 single delay feedback path (R42SDF) architecture adopts a multiplierless radix-4 butterfly structure, based on the specific linear mapping of common factor algorithm (CFA), to support both 256-point fast Fourier transform/inverse fast Fourier transform (FFT/IFFT) and 8 8 2-D discrete cosine transform (DCT) modes following with the high efficient feedback shift registers architecture. The segment shift register (SSR) and overturn shift register (OSR) structure are adopted to minimize the register cost for the input re-ordering and post computation operations in the 8 8 2-D DCT mode, respectively. Moreover, the retrenched constant multiplier and eight-folded complex multiplier structures are adopted to decrease the multiplier cost and the coefficient ROM size with the complex conjugate symmetry rule and subexpression elimination technology. To further decrease the chip cost, a finite wordlength analysis is provided to indicate that the proposed architecture only requires a 13-bit internal wordlength to achieve 40-dB signal-to-noise ratio (SNR) performance in 256-point FFT/IFFT modes and high digital video (DV) compression quality in 8 8 2-D DCT mode. The comprehensive comparison results indicate that the proposed cost effective reconfigurable design has the smallest hardware requirement and largest hardware utilization among the tested architectures for the FFT/IFFT computation, and thus has the highest cost efficiency. The derivation and chip implementation results show that the proposed pipeline 256-point FFT/LFFT/2-D DCT triple-mode chip consumes 22.37 mW at 100 MHz at 1.2-V supply voltage in TSMC 0.13-m CMOS process, which is very appropriate for the RSoCs IP of next-generation handheld devices. &copy; 2008 IEEE.
Lin, C.T., Yang, C.T. & Su, M.T. 2008, 'A hybridization of immune algorithm with particle swarm optimization for neuro-fuzzy classifiers', International Journal of Fuzzy Systems, vol. 10, no. 3, pp. 139-149.
In order to enhance the immune algorithm (IA) performance and find the optimal solution when dealing with difficult problems, we propose an efficient immune-based particle swarm optimization (IPSO) for neuro-fuzzy classifiers to solve the skin color detection problem. The proposed IPSO combines the immune algorithm (IA) and particle swarm optimization (PSO) to perform parameter learning. The IA uses the clonal selection principle, such that antibodies between others of high similar degree are affected, and these antibodies, after the process, will have higher quality, accelerating the search and increasing the global search capacity. The PSO algorithm has proved to be very effective for solving global optimization. It is not only a recently invented high-performance optimizer that is easy to understand and implement, but it also requires little computational bookkeeping and generally only a few lines of code. Hence, we employed the advantages of PSO to improve the mutation mechanism of the immune algorithm. Simulations have shown the performance and applicability of the proposed method. &copy; 2008 TFSA.
Lin, C.-.T., Chuang, S.-.W., Chen, Y.-.C., Ko, L.-.W., Liang, S.-.F. & Jung, T.-.P. 2007, 'EEG effects of motion sickness induced in a dynamic virtual reality environment.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, vol. 2007, pp. 3872-3875.
View/Download from: Publisher's site
The Electroencephalogram (EEG) dynamics which relate to motion sickness are studied in this paper. Instead of providing visual or motion stimuli to the subjects to induce motion sickness, we employed a dynamic virtual-reality (VR) environment in our research. The environment consisted of a 3D surrounding VR scene and a motion platform providing a realistic situation. This environment provided the advantages of safety, low cost, and the realistic stimuli to induce motion sickness. The Motion Sickness Questionnaire (MSQ) was used to assess the sickness level, and the EEG effects on the subjects with high sickness levels were investigated using the independent component analysis (ICA). The fake-epoch extraction was then applied to the nausea-related independent components. Finally we employed the Event-Related Spectral Perturbation (ERSP) technology on the fake-epochs in order to determine the EEG dynamics during motion sickness. The experimental results show that most subjects experienced an 8-10 Hz power increase to their motion sickness-related phenomena in the parietal and motor areas. Moreover, some subjects experienced an EEG power increase of 18-20 Hz in their synchronized responses recorded in the same areas. The motion sickness-related effects and regions can be successfully obtained from our experimental results.
Lin, C.T., Huang, C.H. & Chen, S.A. 2007, 'CNN-based hybrid-order texture segregation as early vision processing and its implementation on CNN-UM', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 54, no. 10, pp. 2277-2287.
View/Download from: Publisher's site
In this paper, a biologically inspired, CNN-based, multi-channel, texture boundary detection technique is presented. The proposed approach is similar to human vision system. The algorithm is simple and straightforward such that it can be implemented on the cellular neural networks (CNNs). CNN contains several important advantages, such as efficient real-time processing capability and feasible very large-scale integration (VLSI) implementation. The proposed algorithm also had been widely tested on synthetic texture images. Those texture images are randomly selected from the Brodatz textures database. According to our simulation results, the boundaries of uniform textures can be detected quite successfully. For the nonuniform or nonregular textures, the results also indicate meaningful properties, and the properties also are consistent to the human visual sensation. The proposed algorithm also has been implemented on the CNN Universal Machine (CNN-UM), and yields similar results as the simulation on the PC. Based on the efficient performance of CNN-UM, the algorithm becomes very fast. &copy; 2007 IEEE.
Lin, C.T., Fan, K.W., Pu, H.C., Lu, S.M. & Liang, S.F. 2007, 'An HVS-directed neural-network-based image resolution enhancement scheme for image resizing', IEEE Transactions on Fuzzy Systems, vol. 15, no. 4, pp. 605-615.
View/Download from: Publisher's site
In this paper, a novel human visual system (HVS)-directed neural-network-based adaptive interpolation scheme for natural image is proposed. A fuzzy decision system built from the characteristics of the HVS is proposed to classify pixels of the input image into human perception nonsensitive class and sensitive class. Bilinear interpolation is used to interpolate the nonsensitive regions and a neural network is proposed to interpolate the sensitive regions along edge directions. High-resolution digital images along with supervised learning algorithms are used to automatically train the proposed neural network. Simulation results demonstrate that the proposed new resolution enhancement algorithm can produce a higher visual quality for the interpolated image than the conventional interpolation methods. &copy; 2007 IEEE.
Lin, K.-.L., Lin, C.-.Y., Huang, C.-.D., Chang, H.-.M., Yang, C.-.Y., Lin, C.-.T., Tang, C.Y. & Hsu, D.F. 2007, 'Feature selection and combination criteria for improving accuracy in protein structure prediction.', IEEE transactions on nanobioscience, vol. 6, no. 2, pp. 186-196.
View/Download from: Publisher's site
The classification of protein structures is essential for their function determination in bioinformatics. At present, a reasonably high rate of prediction accuracy has been achieved in classifying proteins into four classes in the SCOP database according to their primary amino acid sequences. However, for further classification into fine-grained folding categories, especially when the number of possible folding patterns as those defined in the SCOP database is large, it is still quite a challenge. In our previous work, we have proposed a two-level classification strategy called hierarchical learning architecture (HLA) using neural networks and two indirect coding features to differentiate proteins according to their classes and folding patterns, which achieved an accuracy rate of 65.5%. In this paper, we use a combinatorial fusion technique to facilitate feature selection and combination for improving predictive accuracy in protein structure classification. When applying various criteria in combinatorial fusion to the protein fold prediction approach using neural networks with HLA and the radial basis function network (RBFN), the resulting classification has an overall prediction accuracy rate of 87% for four classes and 69.6% for 27 folding categories. These rates are significantly higher than the accuracy rate of 56.5% previously obtained by Ding and Dubchak. Our results demonstrate that data fusion is a viable method for feature selection and combination in the prediction and classification of protein structure.
Hsu, S.C., Liang, S.F., Fan, K.W. & Lin, C.T. 2007, 'A robust in-car digital image stabilization technique', IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews, vol. 37, no. 2, pp. 234-247.
View/Download from: Publisher's site
Machine vision is a key technology used in an intelligent transportation system (ITS) to augment human drivers' visual capabilities. For the in-car applications, additional motion components are usually induced by disturbances such as the bumpy ride of the vehicle or the steering effect, and they will affect the image interpretation processes that is required by the motion field (motion vector) detection in the image. In this paper, a novel robust in-car digital image stabilization (DIS) technique is proposed to stably remove the unwanted shaking phenomena in the image sequences captured by in-car video cameras without the influence caused by moving object (front vehicles) in the image or intentional motion of the car, etc. In the motion estimation process, the representative point matching (RPM) module combined with the inverse triangle method is used to determine and extract reliable motion vectors in plain images that lack features or contain a large low-contrast area to increase the robustness in different imaging conditions, since most of the images captured by in-car video cameras include large low-contrast sky areas. An adaptive background evaluation model is developed to deal with irregular images that contain large moving objects (front vehicles) or a low-contrast area above the skyline. In the motion compensation processing, a compensating motion vector (CMV) estimation method with an inner feedback-loop integrator is proposed to stably remove the unwanted shaking phenomena in the images without losing the effective area of the images with a constant motion condition. The proposed DIS technique was applied to the on-road captured video sequences with various irregular conditions for performance demonstrations. &copy; 2007 IEEE.
Lin, C.-.T., Chung, I.-.F., Ko, L.-.W., Chen, Y.-.C., Liang, S.-.F. & Duann, J.-.R. 2007, 'EEG-based assessment of driver cognitive responses in a dynamic virtual-reality driving environment.', IEEE transactions on bio-medical engineering, vol. 54, no. 7, pp. 1349-1352.
View/Download from: Publisher's site
Accidents caused by errors and failures in human performance among traffic fatalities have a high death rate and become an important issue in public security. They are mainly caused by the failures of the drivers to perceive the changes of the traffic lights or the unexpected conditions happening accidentally on the roads. In this paper, we devised a quantitative analysis for assessing driver's cognitive responses by investigating the neurobiological information underlying electroencephalographic (EEG) brain dynamics in traffic-light experiments in a virtual-reality (VR) dynamic driving environment. The VR technique allows subjects to interact directly with the moving virtual environment instead of monotonic auditory and visual stimuli, thereby provides interactive and realistic tasks without the risk of operating on an actual machine. Independent component analysis (ICA) is used to separate and extract noise-free ERP signals from the multi-channel EEG signals. A temporal filter is used to solve the time-alignment problem of ERP features and principle component analysis (PCA) is used to reduce feature dimensions. The dimension-reduced features are then input to a self-constructing neural fuzzy inference network (SONFIN) to recognize different brain potentials stimulated by red/green/yellow traffic events, the accuracy can be reached 87% in average eight subjects in this visual-stimuli ERP experiment. It demonstrates the feasibility of detecting and analyzing multiple streams of ERP signals that represent operators' cognitive states and responses to task events.
Huang, C.H. & Lin, C.T. 2007, 'Bio-inspired computer fovea model based on hexagonal-type cellular neural network', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 54, no. 1, pp. 35-47.
View/Download from: Publisher's site
For decades, numerous scientists have examined the following questions: "How do humans see the world?"and "How do humans experience vision?"To answer these questions, this study proposes a computer fovea model based on hexagonal-type cellular neural network (hCNN). Certain biological mechanisms of a retina can be simulated using an in-state-of-art architecture named CNN. Those biological mechanisms include the behaviors of the photoreceptors, horizontal cells, ganglions, and bipolar cells, and their co-operations in the retina. Through investigating the model and the abilities of the CNN, various properties of the human vision system can be simulated. The human visual system possesses numerous interesting properties, which provide natural methods of enhancing visual information. Various visual information enhancing algorithms can be developed using these properties and the proposed model. The proposed algorithms include color constancy, image sharpness, and some others. This study also discusses how the proposed model works for video enhancement and demonstrates it experimentally. &copy; 2007 IEEE.
Lin, C.T., Ko, L.W., Chung, I.F., Huang, T.Y., Chen, Y.C., Jung, T.P. & Liang, S.F. 2006, 'Adaptive EEG-based alertness estimation system by using ICA-based fuzzy neural networks', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 53, no. 11, pp. 2469-2476.
View/Download from: Publisher's site
Drivers' fatigue has been implicated as a causal factor in many accidents. The development of human cognitive state monitoring system for the drivers to prevent accidents behind the steering wheel has become a major focus in the field of safety driving. It requires a technique that can continuously monitor and estimate the alertness level of drivers. The difficulties in developing such a system are lack of significant index for detecting drowsiness and the interference of the complicated noise in a realistic and dynamic driving environment. An adaptive alertness estimation methodology based on electroencephalogram, power spectrum analysis, independent component analysis (ICA), and fuzzy neural network (FNNs) models is proposed in this paper for continuously monitoring driver's drowsiness level with concurrent changes in the alertness level. A novel adaptive feature selection mechanism is developed for automatically selecting effective frequency bands of ICA components for realizing an on-line alertness monitoring system based on the correlation analysis between the time-frequency power spectra of ICA components and the driving errors defined as the deviation between the center of the vehicle and the cruising lane in the virtual-reality driving environment. The mechanism also provides effective and efficient features that can be fed into ICA-mixture-model-based self-constructing FNN to indirectly estimate driver's drowsiness level expressed by approximately and predicting the driving error. &copy; 2006 IEEE.
Lin, C.T., Yeh, C.M., Liang, S.F., Chung, J.F. & Kumar, N. 2006, 'Support-vector-based fuzzy neural network for pattern classification', IEEE Transactions on Fuzzy Systems, vol. 14, no. 1, pp. 31-41.
View/Download from: Publisher's site
Fuzzy neural networks (FNNs) for pattern classification usually use the backpropagation or C-cluster type learning algorithms to learn the parameters of the fuzzy rules and membership functions from the training data. However, such kinds of learning algorithms usually cannot minimize the empirical risk (training error) and expected risk (testing error) simultaneously, and thus cannot reach a good classification performance in the testing phase. To tackle this drawback, a support-vector-based fuzzy neural network (SVFNN) is proposed for pattern classification in this paper. The SVFNN combines the superior classification power of support vector machine (SVM) in high dimensional data spaces and the efficient human-like reasoning of FNN in handling uncertainty information. A learning algorithm consisting of three learning phases is developed to construct the SVFNN and train its parameters. In the first phase, the fuzzy rules and membership functions are automatically determined by the clustering principle. In the second phase, the parameters of FNN are calculated by the SVM with the proposed adaptive fuzzy kernel function. In the third phase, the relevant fuzzy rules are selected by the proposed reducing fuzzy rule method. To investigate the effectiveness of the proposed SVFNN classification, it is applied to the Iris, Vehicle, Dna, Satimage, Ijcnn1 datasets from the UCI Repository, Statlog collection and IJCNN challenge 2001, respectively. Experimental results show that the proposed SVFNN for pattern classification can achieve good classification performance with drastically reduced number of fuzzy kernel functions. &copy; 2006 IEEE.
Kau, L.J., Lin, Y.P. & Lin, C.T. 2006, 'Lossless image coding using adaptive, switching algorithm with automatic fuzzy context modelling', IEE Proceedings: Vision, Image and Signal Processing, vol. 153, no. 5, pp. 684-694.
View/Download from: Publisher's site
A switching adaptive predictor (SWAP) with automatic fuzzy context modelling is proposed for lossless image coding. Depending on the context of the coding pixel, the SWAP encoder switches between two predictors: the adaptive neural predictor (ANP) and the texture context matching (TCM) predictor. The ANP is known to perform well and gives small prediction errors except for pixels around edges. For areas with edges, TCM is used. To decide which is to be used, a switching criterion is proposed to pick out pixels around edges effectively. With the switching predictor structure, small prediction errors can be achieved in both slowly varying areas and edges. Furthermore, the use of the so-called fuzzy context clustering for prediction error refinement is proposed. The proposed compensation mechanism is proved to be very useful through experiments. It further improves the bit rates by, on average, 0.2bpp in test images. The experiments also show that an average improvement of 0.3 and 0.05bpp in first-order entropy can be achieved when the proposed switching predictor is compared with the gradient adjusted predictor and a six-order edge directed predictor, respectively. Moreover, the lossless image coder built upon the proposed algorithm also provides lower bit rates than the state-of-the-art context-based, adaptive, lossless image coding (CALIC) system and is comparable to that obtained by the highly complex two-pass coder called TMW. &copy; The Institution of Engineering and Technology 2006.
Lin, C.T., Fan, K.W., Yeh, C.M., Pu, H.C. & Wu, F.Y. 2006, 'High-accuracy skew estimation of document images', International Journal of Fuzzy Systems, vol. 8, no. 3, pp. 119-126.
This paper presents a new skew angle estimation algorithm for binary document images based on the FCRM (fuzzy c-regression models) clustering method with the aim to resolve the disadvantages of low accuracy and robustness of the existing approaches. This algorithm consists of four processes. The first process transfers the input image into parallel straight lines through image analysis. The second process uses the operating window selection to accelerate the executing time. The following process magnifies the image by fast interpolation to increase the accuracy of skew angle estimation. Finally, the FCRM method is applied to estimate the skew angle. A test set of 184 documents of different kinds is used to measure the performance of the proposed algorithm. Experimental results show that the proposed method has a high precision rate for different document types; it is able to accurately estimate the skew angles that range between -89&deg; and +89&deg;. &copy; 2006 TFSA.
Chang, C.L., Fan, K.W., Chung, I.F. & Lin, C.T. 2006, 'A recurrent fuzzy coupled cellular neural network system with automatic structure and template learning', IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 53, no. 8, pp. 602-606.
View/Download from: Publisher's site
The cellular neural network (CNN) is a powerful technique to mimic the local function of biological neural circuits, especially the human visual pathway system, for real-time image and video processing. Recently, many studies show that an integrated CNN system can solve more complex high-level intelligent problems. In this brief, we extend our previously proposed multi-CNN integrated system, called recurrent fuzzy CNN (RFCNN) which considers uncoupled CNNs only, to automatically learn the proper network structure and parameters simultaneously of coupled CNNs, which is called recurrent fuzzy coupled CNN (RFCCNN). The proposed RFCCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. For comparison, the capability of the proposed RFCCNN is demonstrated on the same defect inspection problems. Simulation results show that the proposed RFCCNN outperforms the RFCNN. &copy; 2006 IEEE.
Liang, S.F., Lin, C.T., Wu, R.C., Chen, Y.C., Huang, T.Y. & Jung, T.P. 2005, 'Monitoring driver's alertness based on the driving performance estimation and the EEG power spectrum analysis.', Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, vol. 6, pp. 5738-5741.
Preventing accidents caused by drowsiness behind the steering wheel is highly desirable but requires techniques for continuously estimating driver's abilities of perception, recognition and vehicle control abilities. This paper proposes methods for drowsiness estimation that combine the electroencephalogram (EEG) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate driver's drowsiness level in a virtual-reality-based driving simulator. Results show that it is feasible to quantitatively monitor driver's alertness with concurrent changes in driving performance in a realistic driving simulator.
Lin, C.J. & Chen, C.H. 2005, 'Identification and prediction using recurrent compensatory neuro-fuzzy systems', Fuzzy Sets and Systems, vol. 150, no. 2, pp. 307-330.
View/Download from: Publisher's site
In this paper, a recurrent compensatory neuro-fuzzy system (RCNFS) for identification and prediction is proposed. The compensatory-based fuzzy method uses the adaptive fuzzy operations of neuro-fuzzy systems to make fuzzy logic systems more adaptive and effective. A recurrent network is embedded in the RCNFS by adding feedback connections in the second layer, where the feedback units act as memory elements. In this paper, the RCNFS model is proved to be a universal approximator. Also, an online learning algorithm is proposed which can automatically construct the RCNFS. There are no rules initially in the RCNFS. They are created and adapted as online learning proceeds through simultaneous structure and parameter learning. Structure learning is based on the degree measure and parameter learning is based on the ordered derivative algorithm. Finally, the RCNFS is used in several simulations. The simulation results of the dynamic system model have shown that (1) the RCNFS model converges quickly; (2) the RCNFS model requires a small number of tuning parameters; (3) the RCNFS model can solve temporal problems and approximate a dynamic system. &copy; 2004 Elsevier B.V. All rights reserved.
Hsu, S.C., Liang, S.F. & Lin, C.T. 2005, 'A robust digital image stabilization technique based on inverse triangle method and background detection', IEEE Transactions on Consumer Electronics, vol. 51, no. 2, pp. 335-345.
View/Download from: Publisher's site
In this paper, a novel robust digital image stabilization (DIS) technique is proposed to stably remove the unwanted shaking phenomena in the image sequences captured by hand-held camcorders without affecting moving objects in the image sequence and intentional motion of panning condition, etc. It consists of a motion estimation unit and a motion compensation unit. To increase the robustness in adverse image conditions, an inverse triangle method is proposed to extract reliable motion vectors in plain images which are lack of features or contain large low-contrast area, etc., and a background evaluation model is developed to deal with irregular images which contain large moving objects, etc. In the motion compensation unit, a CMV estimation method with an inner feedback-loop integrator is proposed to stably remove the unwanted shaking phenomena without losing the effective area of the image in panning condition. We also propose a smoothness index (SI) to quantitatively evaluate the performances of different image stabilization methods. The experimental results are on-line available to demonstrate the remarkable performance of the proposed DIS technique. &copy; 2005 IEEE.
Chin, C.L.I. & Lin, C.T. 2005, 'Detection and compensation algorithm for backlight images with fuzzy logic and adaptive compensation curve', International Journal of Pattern Recognition and Artificial Intelligence, vol. 19, no. 8, pp. 1041-1057.
View/Download from: Publisher's site
This paper presents a new algorithm for detection and compensation of backlight images. The proposed technique attacks the weakness of the conventional backlight image processing methods such as over-saturation, losing contrast and so on. The proposed algorithm consists of two operation phases: detection and compensation phases. In the detection phase, we use the spatial position characteristic and histogram of backlight image to obtain two image indices, which can determine the backlight degree of an image. Fuzzy logic is then used to integrate these two indices into a final backlight index determining the final backlight degree of an image precisely. Second, in the compensation phase, to solve the over-saturation problem that exists usually in conventional image compensation methods, we propose the adaptive compensation-curve scheme to compensate and enhance the brightness of backlight images. The luminance of a backlight image is adjusted according to the compensation curve, which is adapted dynamically according to the backlight degree indicated by the backlight index estimated in the detection phase. The performance of the proposed technique is tested on 100 backlight images covering various kinds of backlight conditions and degrees. The experimental and comparison results clearly show the superiority of the proposed technique. &copy; World Scientific Publishing Company.
Lin, C.-.T., Cheng, W.-.C. & Liang, S.-.F. 2005, 'A 3-d surface reconstruction approach based on postnonlinear ICA model.', IEEE transactions on neural networks, vol. 16, no. 6, pp. 1638-1650.
View/Download from: Publisher's site
Photometric stereo technique deals with the reconstruction of three-dimensional (3-D) shape of an object by using several images of the same surface taken from the same viewpoint but under illuminations from different directions. In this paper, we propose a new photometric stereo scheme based on a new reflectance model and the postnonlinear (PNL) independent components analysis (ICA) method. The proposed nonlinear reflectance model consists of diffuse components and specular components for modeling the surface reflectance of a stereo object in an image. Unlike the previous approaches, these two components are not separated and processed individually in the proposed model. An unsupervised learning adaptation algorithm is developed to estimate the reflectance model based on image intensities. In this algorithm, the PNL ICA method is used to obtain the surface normal on each point of an image. Then, the 3-D surface model is reconstructed based on the estimated surface normal on each point of image by using the enforcing integrability method. Two experiments are performed to assess the performance of the proposed approach. We test our algorithm on synthetically generated images for the reconstruction of surface of objects and on a number of real images captured from the Yale Face Database B. These testing images contain variability due to illumination and varying albedo in each point of surface of human faces. All the experimental results are compared to those of the existing photometric stereo approaches tested on the same images. The results clearly indicate the superiority of the proposed nonlinear reflectance model over the conventional Lambertian model and the other linear hybrid reflectance model.
Lin, K.L., Lin, C.Y., Huang, C.D., Chang, H.M., Yang, C.Y., Lin, C.T., Tang, C.Y. & Hsu, D.F. 2005, 'Methods of improving protein structure prediction based on HLA neural network and combinatorial fusion analysis', WSEAS Transactions on Information Science and Applications, vol. 2, no. 12, pp. 2146-2153.
The accurate classification of protein structure is critical and essential for protein function determination in Bioinformatics and Proteomics. A reasonably high rate of prediction accuracy for protein structure classification has been achieved recently in coarse-grained protein class assignment according to their primary amino acid sequences, such as classifying proteins into four classes in SCOP. However, it is still quite a challenge for fine-grained protein fold assignment, especially when the number of possible folding patterns as those defined in SCOP is large. In our previous work, hierarchical learning architecture (HLA) neural networks have been used to differentiate proteins according to their classes and folding patterns. A better prediction accuracy rate for 27 folding categories was 65.5% which improves previous results by Ding and Dubchak with 56.5% prediction accuracy rate. The success of the protein structure classification depends heavily on the computational methods used and the features selected. Here combinatorial fusion analysis (CFA) techniques are used to facilitate feature selection and combination for improving prediction accuracy rate of protein structure classification. The resulting classification has an overall prediction accuracy rate of 87.8% for coarse-grained 4 classes and 70.9% for fine-grained 27 folding categories by applying the concept of CFA to our previous work using neural network with the HLA framework. These results are significantly higher than others and our previous work. They further demonstrate that the CFA techniques can greatly enhance the machine learning method (such as NN in the paper) in the protein structure prediction problem.
Lin, C.-.T., Lin, K.-.L., Yang, C.-.H., Chung, I.-.F., Huang, C.-.D. & Yang, Y.-.S. 2005, 'Protein metal binding residue prediction based on neural networks.', International journal of neural systems, vol. 15, no. 1-2, pp. 71-84.
Over one-third of protein structures contain metal ions, which are the necessary elements in life systems. Traditionally, structural biologists were used to investigate properties of metalloproteins (proteins which bind with metal ions) by physical means and interpreting the function formation and reaction mechanism of enzyme by their structures and observations from experiments in vitro. Most of proteins have primary structures (amino acid sequence information) only; however, the 3-dimension structures are not always available. In this paper, a direct analysis method is proposed to predict the protein metal-binding amino acid residues from its sequence information only by neural networks with sliding window-based feature extraction and biological feature encoding techniques. In four major bulk elements (Calcium, Potassium, Magnesium, and Sodium), the metal-binding residues are identified by the proposed method with higher than 90% sensitivity and very good accuracy under 5-fold cross validation. With such promising results, it can be extended and used as a powerful methodology for metal-binding characterization from rapidly increasing protein sequences in the future.
Lin, C.T., Wu, R.C., Jung, T.P., Liang, S.F. & Huang, T.Y. 2005, 'Estimating driving performance based on EEG spectrum analysis', Eurasip Journal on Applied Signal Processing, vol. 2005, no. 19, pp. 3165-3174.
View/Download from: Publisher's site
The growing number of traffic accidents in recent years has become a serious concern to society. Accidents caused by driver's drowsiness behind the steering wheel have a high fatality rate because of the marked decline in the driver's abilities of perception, recognition, and vehicle control abilities while sleepy. Preventing such accidents caused by drowsiness is highly desirable but requires techniques for continuously detecting, estimating, and predicting the level of alertness of drivers and delivering effective feedbacks to maintain their maximum performance. This paper proposes an EEC-based drowsiness estimation system that combines electroencephalogram (EEC) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate driver's drowsiness level in a virtual-reality-based driving simulator. Our results demonstrated that it is feasible to accurately estimate quantitatively driving performance, expressed as deviation between the center of the vehicle and the center of the cruising lane, in a realistic driving simulator. &copy; 2005 Hindawi Publishing Corporation.
Lin, C.T., Cheng, W.C. & Liang, S.F. 2005, 'An on-line ICA-mixture-model-based self-constructing fuzzy neural network', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 52, no. 1, pp. 207-221.
View/Download from: Publisher's site
This paper proposes a new fuzzy neural network (FNN) capable of parameter self-adapting and structure self-constructing to acquire a small number of fuzzy rules for interpreting the embedded knowledge of a system from the given training data set. The proposed FNN is inherently a modified Takagi-Sugeno-Kang (TSK)-type fuzzy-rule-based model with neural network's learning ability. There are no rules initiated at the beginning and they are created and adapted through an on-line learning processing that performs simultaneous structure and parameter identification. In the structure identification of the precondition part, the input space is partitioned in a flexible way according to the newly proposed on-line independent component analysis (ICA) mixture model. The input space is thus represented by linear combinations of independent, non-Gaussian densities. The first input training pattern is assigned to the first rule initially by the on-line ICA mixture model. Afterwards, some additional significant terms (input variables) selected by the on-line ICA mixture model will be added to the consequent part (forming a liner equation of input variables) incrementally or create a new rule in the learning processing. The combined precondition and consequent structure identification scheme can make the network grow dynamically and efficiently. In the parameter identification, the consequent parameters are tuned by the backpropagation rule and the precondition parameters are turned by the on-line ICA mixture model. Both the structure and parameter identifications are done simultaneously to form a fast learning scheme. The derived on-line ICA mixture model also provide a natural linear transformation for each input variable to enhance the knowledge representation ability of the proposed FNN and reduce the required rules and achieve higher accuracy efficiently. In order to demonstrate the performance of the proposed FNN, several experiments covering the areas of system identifica...
Lin, C.T., Wu, R.C., Liang, S.F., Chao, W.H., Chen, Y.J. & Jung, T.P. 2005, 'EEG-based drowsiness estimation for safety driving using independent component analysis', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 52, no. 12, pp. 2726-2738.
View/Download from: Publisher's site
Preventing accidents caused by drowsiness has become a major focus of active safety driving in recent years. It requires an optimal technique to continuously detect drivers' cognitive state related to abilities in perception, recognition, and vehicle control in (near-) real-time. The major challenges in developing such a system include: 1) the lack of significant index for detecting drowsiness and 2) complicated and pervasive noise interferences in a realistic and dynamic driving environment. In this paper, we develop a drowsiness-estimation system based on electroencephalogram (EEG) by combining independent component analysis (ICA), power-spectrum analysis, correlation evaluations, and linear regression model to estimate a driver's cognitive state when he/ she drives a car in a virtual reality (VR)-based dynamic simulator. The driving error is defined as deviations between the center of the vehicle and the center of the cruising lane in the lane-keeping driving task. Experimental results demonstrate the feasibility of quantitatively estimating drowsiness level using ICA-based multistream EEG spectra. The proposed ICA-based method applied to power spectrum of ICA components can successfully (1) remove most of EEG artifacts, (2) suggest an optimal montage to place EEG electrodes, and estimate the driver's drowsiness fluctuation indexed by the driving performance measure. Finally, we present a benchmark study in which the accuracy of ICA-component-based alertness estimates compares favorably to scalp-EEG based. &copy; 2005 IEEE.
Huang, C.D., Liang, S.F.U., Lin, C.T. & Ruei-Cheng, W.U. 2005, 'Machine learning with automatic feature selection for multi-class protein fold classification', Journal of Information Science and Engineering, vol. 21, no. 4, pp. 711-720.
The use of a machine learning approach with automatic feature selection for multi-class protein fold classification is studied. Neural networks are used to complete the task of protein fold classification, where each node is associated with a gate. The results show that the proposed architecture is effective in reducing the dimensionality of the data and enhancing the classification performance. The proposed technique allows the processing of more features from amino acid sequences.
Lin, C.T., Chung, J.F. & Pu, H.C. 2005, 'Pedestrian detection system', International Journal of Fuzzy Systems, vol. 7, no. 2, pp. 45-52.
In this paper, we propose a new pedestrian detection algorithm, and use it to develop a real-time pedestrian detection system. The pedestrian detection algorithm can be functionally partitioned into two parts: moving object segmentation and pedestrian recognition. In moving object segmentation, we segment the moving objects in the scene by modified temporal differencing method. This method combines general temporal differencing method with detection nets. In pedestrian recognition, we obtain multi-type wavelet templates from input images. Interest points are extracted from wavelet templates. Then, feature points are extracted by statistical method from interest point templates. Finally, these feature points are fed into a trained multilayer back-propagation neural network. The output of the neural network implies the result - pedestrian or non-pedestrian. From experiments, accuracy rate of recognition achieves 95%. We implement this algorithm in a real-time pedestrian detection system. The system can detect pedestrians in the scene in real time. The pedestrian detection rate is 92.58%. &copy; 2005 TFSA.
Lin, C.T., Chang, C.L. & Chung, J.F. 2005, 'New horizon for CNN: With fuzzy paradigms for multimedia', IEEE Circuits and Systems Magazine, vol. 5, no. 2, pp. 20-35.
View/Download from: Publisher's site
The cellular neural network (CNN) is a powerful technique to mimic the local function of biological neural circuits for real-time image and video processing. Recently, it is widely accepted that using a set of CNNs in parallel can achieve higher-level information processing and reasoning functions either from application or biologics points of views. We introduce a novel framework for constructing a multiple-CNN integrated neural system called recurrent fuzzy CNN (RFCNN). This system can automatically learn its proper network structure and parameters simultaneously. In the RFCNN, each learned fuzzy rule corresponds to a CNN. Hence, each CNN takes care of a fuzzily separated problem region, and the functions of all CNNs are integrated through the fuzzy inference mechanism. Some on-line clustering algorithms are introduced for the structure learning, and the ordered-derivative calculus is applied to derive the recurrent learning rules of CNN templates in the parameter-learning phase. RFCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. The capability of the RFCNN is demonstrated on the real-world vision-based defect inspection and image descreening problems proving that the RFCNN scheme is effective and promising. &copy;2005 IEEE.
Lin, C.-.T., Cheng, W.-.C. & Liang, S.-.F. 2005, 'Neural-network-Based adaptive hybrid-reflectance model for 3-D surface reconstruction.', IEEE transactions on neural networks, vol. 16, no. 6, pp. 1601-1615.
View/Download from: Publisher's site
This paper proposes a novel neural-network-based adaptive hybrid-reflectance three-dimensional (3-D) surface reconstruction model. The neural network automatically combines the diffuse and specular components into a hybrid model. The proposed model considers the characteristics of each point and the variant albedo to prevent the reconstructed surface from being distorted. The neural network inputs are the pixel values of the two-dimensional images to be reconstructed. The normal vectors of the surface can then be obtained from the output of the neural network after supervised learning, where the illuminant direction does not have to be known in advance. Finally, the obtained normal vectors are applied to enforce integrability when reconstructing 3-D objects. Facial images and images of other general objects were used to test the proposed approach. The experimental results demonstrate that the proposed neural-network-based adaptive hybrid-reflectance model can be successfully applied to objects generally, and perform 3-D surface reconstruction better than some existing approaches.
Song, C., Chai, T., Wu, S.J. & Lin, C.T. 2005, 'Comment on "Discrete-time optimal fuzzy controller design: Global concept approach"', IEEE Transactions on Fuzzy Systems, vol. 13, no. 2, pp. 285-286.
View/Download from: Publisher's site
In this correspondence, we show that the relationship between the finite optimal trajectory and the infinite optimal trajectory in Lemma 3 of the aforementioned paper cannot act as a theory basis for the further implementation of other theorems and the dynamic decomposition algorithm. &copy; 2005 IEEE.
Lin, C.T., Wu, R.C., Chang, J.Y. & Liang, S.F. 2004, 'A novel prosodic-information synthesizer based on recurrent fuzzy neural network for the Chinese TTS system.', IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society, vol. 34, no. 1, pp. 309-324.
In this paper, a new technique for the Chinese text-to-speech (TTS) system is proposed. Our major effort focuses on the prosodic information generation. New methodologies for constructing fuzzy rules in a prosodic model simulating human's pronouncing rules are developed. The proposed Recurrent Fuzzy Neural Network (RFNN) is a multilayer recurrent neural network (RNN) which integrates a Self-cOnstructing Neural Fuzzy Inference Network (SONFIN) into a recurrent connectionist structure. The RFNN can be functionally divided into two parts. The first part adopts the SONFIN as a prosodic model to explore the relationship between high-level linguistic features and prosodic information based on fuzzy inference rules. As compared to conventional neural networks, the SONFIN can always construct itself with an economic network size in high learning speed. The second part employs a five-layer network to generate all prosodic parameters by directly using the prosodic fuzzy rules inferred from the first part as well as other important features of syllables. The TTS system combined with the proposed method can behave not only sandhi rules but also the other prosodic phenomena existing in the traditional TTS systems. Moreover, the proposed scheme can even find out some new rules about prosodic phrase structure. The performance of the proposed RFNN-based prosodic model is verified by imbedding it into a Chinese TTS system with a Chinese monosyllable database based on the time-domain pitch synchronous overlap add (TD-PSOLA) method. Our experimental results show that the proposed RFNN can generate proper prosodic parameters including pitch means, pitch shapes, maximum energy levels, syllable duration, and pause duration. Some synthetic sounds are online available for demonstration.
Shou, Y.W. & Lin, C.T. 2004, 'Image descreening by GA-CNN-based texture classification', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 51, no. 11, pp. 2287-2299.
View/Download from: Publisher's site
This paper proposes a new image-descreening technique based on texture classification using a cellular neural network (CNN) with template trained by genetic algorithm (GA), called GA-CNN. Instead of using the fixed filters for image descreening, we are equipped with a more pliable mechanism for classifications in screening patterns. Using CNN makes it possible to get an accurate texture classification result in a faster speed by its superiority of implementable hardware and the flexible choices of templates. The use of the GA here helps us to look for the most appropriate template for CNNs more adaptively and methodically. The evolved parameters in the template for CNNs can not only provide a quicker classification mechanism but also help us with a better texture classification for screening patterns. After the class of screening patterns in the querying images is determined by the trained GA-CNN-based texture classification system, the recommendatory filters are induced to solve the screening problems. The induction of the classification in screening patterns has simplified the choice of filters and made it valueless to determine a new structured filter. Eventually, our comprehensive methodology is going to be topped off with more desirable results and the indication for the decrease in time complexity. &copy; 2004 IEEE.
Duh, F.B., Juang, C.F. & Lin, C.T. 2004, 'A Neural Fuzzy Network Approach to Radar Pulse Compression', IEEE Geoscience and Remote Sensing Letters, vol. 1, no. 1, pp. 15-20.
View/Download from: Publisher's site
To make good range resolution and accuracy compatible with a high detection capability while maintaining the low average transmitted power, pulse compression processing giving low-range sidelobes is necessary. The traditional algorithms such as the direct autocorrelation filter (ACF), least squares (LS) inverse filter, and linear programming (LP) filter based on three-element Barker code (B13 code) have been developed. Recently, the neural network algorithms were issued. However, the traditional algorithms cannot achieve the requirements of high signal-to-sidelobe ratio and low integrated sidelobe level (ISL), and the normal neural networks such as the backpropagation (BP) network usually produce the extra problems of low convergence speed and are sensitive to the Doppler frequency shift. To overcome these defects, a new approach using a neural fuzzy network to deal with pulse compression in a radar system is presented. Two different Barker codes are carried out by a six-layer self-constructing neural fuzzy network (SONFIN). Simulation results show that this neural fuzzy network pulse compression (NFNPC) algorithm has significant advantages in noise rejection performance, range resolution ability, and Doppler tolerance, which are superior to the traditional and BP algorithms.
Huang, C.D. & Lin, C.T. 2004, 'The NN and SVM hierarchical learning architecture for multi-class protein fold recognition', International Journal of Fuzzy Systems, vol. 6, no. 1, pp. 20-27.
The classification of the structure of protein plays a very important role in biological data. It is well known that by means of the classification, the relationships and characteristics among those known proteins can be exploited to predict the structure of new proteins. In general, the study and discovery of the protein structures is based on the sequences and their similarity. In the past, because the complexity of protein sequences, corrective classification is a difficult task. Recently, due to the ability of machine learning techniques, many researchers have applied them to probe into this protein classification problem. Here, we also apply machine-learning methods for multi-class protein fold recognition problem by proposing a novel hierarchical learning architecture. This novel hierarchical learning architecture can be formed by NN (neural networks) or SVM (support vector machine) as basic building blocks. Our results show that, with the help of this novel hierarchical learning architecture, both of NN and SVM can perform well. We use this new architecture to attack the multi-class protein fold recognition problem as proposed by Dubchak and Ding in 2001. With the same set of features and patterns numbers our method can not only obtain better prediction accuracy and lower computation time, moreover, also can avoid the use of the stochastic voting process in the original approach.
Duh, F.B. & Lin, C.T. 2004, 'Tracking a Maneuvering Target Using Neural Fuzzy Network', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 34, no. 1, pp. 16-33.
View/Download from: Publisher's site
A fast target maneuver detecting and highly accurate tracking technique using a neural fuzzy network based on Kalman filter is proposed in this paper. In the automatic target tracking system, there exists an important and difficult problem: how to detect the target maneuvers and fast response to avoid misstracking? The traditional maneuver detection algorithms, such as variable dimension filter (VDF) and input estimation (IE) etc., are computation intensive and difficult to implement in real time. To solve this problem, neural network algorithms have been issued recently. However, the normal neural networks such as backpropagation networks usually produce the extra problems of low convergence speed and/or large network size. Furthermore, the way to decide the network structure is heuristic. To overcome these defects and to make use of neural learning ability, a developed standard Kalman filter with a self-constructing neural fuzzy inference network (KF-SONFIN) algorithm for target tracking is presented in this paper. By generating possible target trajectories including maneuver information to train the SONFIN, the trained SONFIN can detect when the maneuver occurred, the magnitude of maneuver values and when the maneuver disappeared. Without having to change the structure of Kalman filter nor modeling the maneuvering target, this new algorithm, SONFIN, can always find itself an economic network size with a fast learning process. Simulation results show that the KF-SONFIN is superior to the traditional IE and VDF methods in estimation accuracy.
Lin, C.T., Lin, K.L., Yang, C.H., Chung, I.F., Huang, C.D. & Yang, Y.S. 2004, 'Protein metal binding residue prediction based on neural networks', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3316, pp. 1316-1321.
It is known that over one-third of protein structures contain metal ions, and they are the necessary elements in life system. Traditionally, structural biologists used to investigate properties of metalloproteins (proteins which bind with metal ions) by physical means and interpret the function formation and reaction mechanism of enzyme by their structures and observation from experiments in vitro. Most of proteins have primary structures (amino acid sequence information) only; however, the 3-dimension structures are not always available. In this paper, a direct analysis method is proposed to predict protein metalbinding amino acid residues only from its sequence information by neural network with sliding window-based feature extraction and biological feature encoding techniques and it can successfully detect 15 binding elements in protein, and 6 binding elements in enzyme. &copy; Springer-Verlag Berlin Heidelberg 2004.
Lin, C.T., Chang, C.L. & Cheng, W.C. 2004, 'A recurrent fuzzy cellular neural network system with automatic structure and template learning', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 51, no. 5, pp. 1024-1035.
View/Download from: Publisher's site
It is widely accepted that using a set of cellular neural networks (CNNs) in parallel can achieve higher level information processing and reasoning functions either from application or biologics points of views. Such an integrated CNN system can solve more complex intelligent problems. In this paper, we propose a novel framework for automatically constructing a multiple-CNN integrated neural system in the form of a recurrent fuzzy neural network. This system, called recurrent fuzzy CNN (RFCNN), can automatically learn its proper network structure and parameters simultaneously. The structure learning includes the fuzzy division of the problem domain and the creation of fuzzy rules and CNNs. The parameter learning includes the tuning of fuzzy membership functions and CNN templates. In the RFCNN, each learned fuzzy rule corresponds to a CNN. Hence, each CNN takes care of a fuzzily separated problem region, and the functions of all CNNs are integrated through the fuzzy inference mechanism. A new online adaptive independent component analysis mixture-model technique is proposed for the structure learning of RFCNN, and the ordered-derivative calculus is applied to derive the recurrent learning rules of CNN templates in the parameter-learning phase. The proposed RFCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. The capability of the proposed RFCNN is demonstrated on the real-world defect inspection problems. Experimental results show that the proposed scheme is effective and promising.
Lu, U., Hu, B.C.P., Shih, Y.C., Yang, Y.S., Wu, C.Y., Yuan, C.J., Ker, M.D., Wu, T.K., Li, Y.K., Hsieh, Y.Z., Hsu, W. & Lin, C.T. 2003, 'CMOS chip as luminescent sensor for biochemical reactions', IEEE Sensors Journal, vol. 3, no. 3, pp. 310-316.
View/Download from: Publisher's site
We describe a novel biochemical sensing method and its potential new biosensing applications. A light-sensitive complementary metal oxide semiconductor (CMOS) chip prepared through a standard 0.5-m CMOS process was developed for measuring biochemical reactions. A light producing enzymatic reaction catalyzed by horseradish peroxidase (HRP) was designed as a platform reaction to determine the concentration of hydrogen peroxide (H 2O2) by the CMOS chip with a standard semiconductor parameter analyzer (HP4145). The kinetics of enzymatic reaction were determined and compared with a standard and sophisticated fluorometer (Hitachi F-4500) in a biochemical laboratory. Similar results were obtained by both instruments. Using glucose oxidase as an example, we further demonstrated that the HRP platform can be used to determine other H2O2 producing reactions with the CMOS system. The result points to an important application of the CMOS chip in biological measurements and in diagnosis of various health factors.
Lin, C.-.T., Liu, S.-.H., Wang, J.-.J. & Wen, Z.-.C. 2003, 'Reduction of interference in oscillometric arterial blood pressure measurement using fuzzy logic.', IEEE transactions on bio-medical engineering, vol. 50, no. 4, pp. 432-441.
View/Download from: Publisher's site
In oscillometry, oscillation amplitudes (OAs) embedded in the cuff pressure are drastically affected by a variety of artifacts and cardiovascular diseases, leading to inaccurate arterial blood pressure (ABP) measurement. The purpose of this paper is to improve the accuracy in the arterial pressure measurement by reducing interference in the OAs using a recursive weighted regression algorithm (RWRA). This method includes a fuzzy logic discriminator (FLD) and a recursive regression algorithm. The FLD is used to reduce the effect of artifacts caused by measurement motion disturbance or cardiovascular diseases, and to determine the truthfulness of the oscillation pulse. According to the truth degree, the relationship between the cuff pressure and OA is reconstructed using the regression algorithm. Because the regression method must utilize inverse matrix operation, which will be difficult to implement in an automatic or ambulatory monitor, the recursive regression method is proposed to solve this problem. To test the performance of this RWRA, 47 subjects underwent the ABP measurement using both the auscultation and the oscillometry combined with the RWRA. It was found that the average difference between the pooled blood pressures measured by the auscultation and those by the oscillometry combined with the RWRA was found to be only 4.9 mmHg. Clinical results demonstrated that the proposed RWRA is more robust than the traditional curve fitting algorithm (TCFA). We conclude that the proposed RWRA can be applied to effectively improve the accuracy of the oscillometric blood pressure measurement.
Huang, C.D., Chung, I.F., Pal, N.R. & Lin, C.T. 2003, 'Machine learning for multi-class protein fold classification based on neural networks with feature gating', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2714, pp. 1168-1175.
The success of a classification system depends heavily on two things: the tools being used and the features considered. For the bioinformatics applications the role of appropriate features has not been paid adequate importance. In this investigation we use two novel ideas. First, we use neural networks where each input node is associated with a gate. At the beginning of the training all gates are almost closed, i.e., no feature is allowed to enter the network. During the training, depending on the requirements, gates are either opened or closed. At the end of the training, gates corresponding to good features are completely opened while gates corresponding to bad features are closed more tightly. And of course, some gates may be partially open. So the network can not only select features in an online manner when the learning goes on, it also does some feature extraction. The second novel idea is to use a hierarchical machine learning architecture. Where at the first level the network classifies the data into four major folds : all alpha, all beta, alpha + beta and alpha / beta. And in the next level we have another set of networks, which further classifies the data into twenty seven folds. This approach helps us to achieve the following. The gating network is found to reduce the number of features drastically. It is interesting to observe that for the first level using just 50 features selected by the gating network we can get a comparable test accuracy as that using 125 features using neural classifiers. The process also helps us to get a better insight into the folding process. For example, tracking the evolution of different gates we can find which characteristics (features) of the data are more important for the folding process. And, of course, it reduces the computation time. The use of the hierarchical architecture helps us to get a better performance also. &copy; Springer-Verlag Berlin Heidelberg 2003.
Huang, C.-.D., Lin, C.-.T. & Pal, N.R. 2003, 'Hierarchical learning architecture with automatic feature selection for multiclass protein fold classification.', IEEE transactions on nanobioscience, vol. 2, no. 4, pp. 221-232.
View/Download from: Publisher's site
The structure classification of proteins plays a very important role in bioinformatics, since the relationships and characteristics among those known proteins can be exploited to predict the structure of new proteins. The success of a classification system depends heavily on two things: the tools being used and the features considered. For the bioinformatics applications, the role of appropriate features has not been paid adequate importance. In this investigation we use three novel ideas for multiclass protein fold classification. First, we use the gating neural network, where each input node is associated with a gate. This network can select important features in an online manner when the learning goes on. At the beginning of the training, all gates are almost closed, i.e., no feature is allowed to enter the network. Through the training, gates corresponding to good features are completely opened while gates corresponding to bad features are closed more tightly, and some gates may be partially open. The second novel idea is to use a hierarchical learning architecture (HLA). The classifier in the first level of HLA classifies the protein features into four major classes: all alpha, all beta, alpha + beta, and alpha/beta. And in the next level we have another set of classifiers, which further classifies the protein features into 27 folds. The third novel idea is to induce the indirect coding features from the amino-acid composition sequence of proteins based on the N-gram concept. This provides us with more representative and discriminative new local features of protein sequences for multiclass protein fold classification. The proposed HLA with new indirect coding features increases the protein fold classification accuracy by about 12%. Moreover, the gating neural network is found to reduce the number of features drastically. Using only half of the original features selected by the gating neural network can reach comparable test accuracy as that using all the ori...
Chung, I.F., Huang, C.D., Shen, Y.H. & Lin, C.T. 2003, 'Recognition of structure classification of protein folding by NN and SVM hierarchical learning architecture', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2714, pp. 1159-1167.
Classifying the structure of protein is a very important task in biological data. By means of the classification, the relationships and characteristics among known proteins can be exploited to predict the structure of new proteins. The study of the protein structures is based on the sequences and their similarity. It is a difficult task. Recently, due to the ability of machine learning techniques, many researchers have applied them to probe into this protein classification problem. We also apply here machine learning methods for multi-class protein fold recognition problem by proposing a novel hierarchical learning architecture. This novel hierarchical learning architecture can be formed by NN (neural networks) or SVM (support vector machine) as basic building blocks. Our results show that both of them can perform well. We use this new architecture to attack the multi-class protein fold recognition problem as proposed by Dubchak and Ding in 2001. With the same set of features our method can not only obtain better prediction accuracy and lower computation time, but also can avoid the use of the stochastic voting process in the original approach. &copy; Springer-Verlag Berlin Heidelberg 2003.
Lin, C.T. 2003, 'Single-Channel Speech Enhancement in Variable Noise-Level Environment', IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans., vol. 33, no. 1, pp. 137-144.
View/Download from: Publisher's site
This paper discusses the problem of single-channel speech enhancement in variable noise-level environment. Commonly used, single-channel subtractive-type speech enhancement algorithms always assume that the background noise level is fixed or slowly varying. In fact, the background noise level may vary quickly. This condition usually results in wrong speech/noise detection and wrong speech enhancement process. In order to solve this problem, we propose a new subtractive-type speech enhancement scheme in this paper. This new enhancement scheme uses the RTF (refined time-frequency parameter)-based RSONFIN (recurrent self-organizing neural fuzzy inference network) algorithm we developed previously to detect the word boundaries in the condition of variable background noise level. In addition, a new parameter (MiFre) is proposed to estimate the varying background noise level. Based on this parameter, the noise level information used for subtractive-type speech enhancement can be estimated not only during speech pauses, but also during speech segments. This new subtractive-type enhancement scheme has been tested and found to perform well, not only in variable background noise level condition, but also in fixed background noise level condition.
Zhou, S., Feng, G., Wu, S.J. & Lin, C.T. 2003, 'Comment on "optimal fuzzy controller design: Local concept approach" (multiple letters)', IEEE Transactions on Fuzzy Systems, vol. 11, no. 2, pp. 279-280.
View/Download from: Publisher's site
Lin, C.T., Chung, I.F., Pu, H.C., Lee, T.H. & Chang, J.Y. 2002, 'Genetic algorithm-based neural fuzzy decision tree for mixed scheduling in ATM networks', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 32, no. 6, pp. 832-845.
View/Download from: Publisher's site
Future broad-band integrated services networks based on the asynchronous transfer mode (ATM) technology are expected to support multiple types of multimedia information with diverse statistical characteristics and quality of service (QoS) requirements. To meet these requirements, efficient scheduling methods are important for traffic control in the ATM networks. Among the general scheduling schemes, the rate monotonic algorithm is simple enough to be used in high-speed networks, but it does not attain as high a system utilization as the deadline driven algorithm does. However, the deadline driven scheme is computationally complex and hard to implement in hardware. The mixed scheduling algorithm is the combination of the rate monotonic algorithm and the deadline driven algorithm; thus it can provide most of the benefits of these two algorithms. In this paper, we use the mixed scheduling algorithm to achieve high system utilization under the hardware constraint. Because there is no analytic method for the schedulability test of the mixed scheduling, we propose a genetic algorithm-based neural fuzzy decision tree (GANFDT) to realize it in a real-time environment. The GANFDT combines the GA and a neural fuzzy network into a binary classification tree. This approach also exploits the power of the classification tree. Simulation results show that the GANFDT provides an efficient way to carry out the mixed scheduling in the ATM networks.
Lin, C.T., Lin, J.Y. & Wu, G.D. 2002, 'A Robust Word Boundary Detection Algorithm for Variable Noise-Level Environment in Cars', IEEE Transactions on Intelligent Transportation Systems, vol. 3, no. 1, pp. 89-100.
View/Download from: Publisher's site
This paper discusses the problem of automatic word boundary detection in the presence of variable-level background noise in cars. Commonly used robust word boundary detection algorithms always assume that the background noise level is fixed and sets fixed thresholds to find the boundary of word signal. In fact, the background noise level in cars varies in the procedure of recording due to speed change and moving environment, and some thresholds should be tuned according to the variation of background noise level. This is the major reason that most robust word boundary detection algorithms cannot work well in the condition of variable background noise level. To solve this problem, we propose a minimum mel-scale frequency band (MiMSB) parameter which can estimate the varying background noise level in cars by adaptively choosing one band with minimum energy, from the mel-scale frequency bank. With the MiMSB parameter, some preset thresholds used to find the boundary of word signal are no longer fixed in all the recording intervals. These thresholds will be tuned according to the MiMSB parameter. We also propose an enhanced time-frequency (ETF) parameter by extending the time-frequency (TF) parameter proposed by Junqua et al. from single band to multiband spectrum analysis, where the frequency bands help to make the distinction between speech signal and noise. The ETF parameter can extract useful frequency information by choosing some bands of the mel-scale frequency bank. Based on the MiMSB and ETF parameters, we finally propose a new robust algorithm for word boundary detection in variable noise-level environment. The new algorithm has been tested over a variety of noise conditions in cars and has been found to perform well not only under variable background noise level condition, but also under fixed background noise level condition. The new robust algorithm using the MiMSB and ETF parameters achieved higher recognition rate than the TF-based robust algorithm, whi...
Lin, C.T., Wu, R.C. & Wu, G.D. 2002, 'Noisy speech segmentation/enhancement with multiband analysis and neural fuzzy networks', International Journal of Pattern Recognition and Artificial Intelligence, vol. 16, no. 7, pp. 927-955.
View/Download from: Publisher's site
This paper addresses the problem of speech segmentation and enhancement in the presence of noise. We first propose a new word boundary detection algorithm by using a neural fuzzy network (called ATF-based SONFIN algorithm) for identifying islands of word signals in fixed noise-level environment. We further propose a new RTF-based RSONFIN algorithm where the background noise level varies during the procedure of recording. The adaptive time-frequency (ATF) and refined time-frequency (RTF) parameters extend the TF parameter from single band to multiband spectrum analysis, and help to make the distinction of speech and noise signals clear. The ATF and RTF parameters can extract useful frequency information by adaptively choosing proper bands of the mel-scale frequency bank. Due to the self-learning ability of SONFIN and RSONFIN, the proposed algorithms avoid the need of empirically determining thresholds and ambiguous rules. The RTF-based RSONFIN algorithm can also find the variation of the background noise level and detect correct word boundaries in the condition of variable background noise level by processing the temporal relations. Our experimental results show that both in the fixed and variable noise-level environment, the algorithms that we proposed achieved higher recognition rate than several commonly used word boundary detection algorithms and reduced the recognition error rate due to endpoint detection.
Wu, S.J. & Lin, C.T. 2002, 'Global optimal fuzzy tracker design based on local concept approach', IEEE Transactions on Fuzzy Systems, vol. 10, no. 2, pp. 128-143.
View/Download from: Publisher's site
In this paper, we propose a global optimal fuzzy tracking controller, implemented by fuzzily blending the individual local fuzzy tracking laws, for continuous and discrete-time fuzzy systems with the aim of solving, respectively, the continuous and discrete-time quadratic tracking problems with moving or model-following targets under finite or infinite horizon (time). The differential or recursive Riccati equations, and more, the differential or difference equations in tracing the variation of the target are derived. Moreover, in the case of time-invariant fuzzy tracking systems, we show that the optimal tracking controller can be obtained by just solving algebraic Riccati equations and algebraic matrix equations. Grounding on this, several fascinating characteristics of the resultant closed-loop continuous or discrete time-invariant fuzzy tracking systems can be elicited easily. The stability of both closed-loop fuzzy tracking systems can be ensured by the designed optimal fuzzy tracking controllers. The optimal closed-loop fuzzy tracking systems cannot only be guaranteed to be exponentially stable, but also be stabilized to any desired degree. Moreover, the resulting closed-loop fuzzy tracking systems possess infinite gain margin; that is, their stability is guaranteed no matter how large the feedback gain becomes. Two examples are given to illustrate the performance of the proposed optimal fuzzy tracker design schemes and to demonstrate the proved stability properties.
Wang, J.J., Lin, C.T., Liu, S.H. & Wen, Z.C. 2002, 'Model-based synthetic fuzzy logic controller for indirect blood pressure measurement', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 32, no. 3, pp. 306-315.
View/Download from: Publisher's site
In this paper, a new measurement system for the non-invasive monitoring of the continuous blood pressure waveform in the radial artery is presented. The proposed system comprises a model-based fuzzy logic controller, an arterial tonometer and a micro syringe device. The flexible diaphragm tonometer is to register the continuous blood pressure waveform. To obtain accurate measurement without distortion, the tonometer's mean chamber pressure must be kept equal to the mean arterial pressure (MAP), the so-called optimal coupling condition, such that the arterial vessel has the maximum compliance. Since the MAP cannot be measured directly, to keep the optimal coupling condition becomes a tracking control problem with unknown desired trajectory. To solve this dilemma, a model-based fuzzy logic controller is designed to compensate the change of MAP by applying a counter pressure on the tonometer chamber through the micro syringe device. The proposed controller consists of a model-based predictor and a synthetic fuzzy logic controller (SFLC). The model-based predictor is to estimate the MAPs changing tendency based on the identified arterial pressure-volume model. The SFLC is composed of three subcontrollers, each of which is a simple fuzzy logic controller, for processing the three changing states of the MAP: ascending, descending and stabilizing states, respectively. Simulation results show that, for the MAP with changing rates of &plusmn;10, &plusmn;20 or &plusmn;30 mm Hg/min, the model-based SFLC can beat-to-beat adjust the tonometer's chamber pressure only with a mean square error of 1.9, 2.2, or 2.8 mm Hg, respectively.
Shich, C.S. & Lin, C.T. 2002, 'A vector neural network for emitter identification', IEEE Transactions on Antennas and Propagation, vol. 50, no. 8, pp. 1120-1127.
View/Download from: Publisher's site
This paper proposes a three-layer vector neural network (VNN) with a supervised learning algorithm suitable for signal classification in general, and for emitter identification (EID) in particular. The VNN can accept interval-value input data as well as scalar input data. The input features of the EID problems include the radio frequency, pulse width, and pulse repetition interval of a received emitter signal. Since the values of these features vary in interval ranges in accordance with a specific radar emitter, the VNN is proposed to process interval-value data in the EID problem. In the training phase, the interval values of the three features are presented to the input nodes of VNN. A new vector-type backpropagation learning algorithm is derived from an error function defined by the VNN's actual output and the desired output indicating the correct emitter type of the corresponding feature intervals. The algorithm can tune the weights of VNN optimally to approximate the nonlinear mapping between a given training set of feature intervals and the corresponding set of desired emitter types. After training, the VNN can be used to identify the sensed scalar-value features from a real-time received emitter signal. A number of simulations are presented to demonstrate the effectiveness and identification capability of VNN, including the two-EID problem and the multi-EID problem with/without additive noise. The simulated results show that the proposed algorithm cannot only accelerate the convergence speed, but it can help avoid getting stuck in bad local minima and achieve higher classification rate.
Wu, S.J. & Lin, C.T. 2002, 'Discrete-time optimal fuzzy controller design: Global concept approach', IEEE Transactions on Fuzzy Systems, vol. 10, no. 1, pp. 21-38.
View/Download from: Publisher's site
In this paper, we propose a systematic and theoretically sound way to design a global optimal discrete-time fuzzy controller to control and stabilize a nonlinear discrete-time fuzzy system with finite or infinite horizon (time). A linear-like global system representation of discrete-time fuzzy system is first proposed by viewing a discrete-time fuzzy system in global concept and unifying the individual matrices into synthetical matrices. Then, based on this kind of system representation, the discrete-time optimal fuzzy control law which can achieve global minimum effect is developed theoretically. A nonlinear two-point-boundary-value-problem (TPBVP) is derived as the necessary and sufficient condition for the nonlinear quadratic optimal control problem. To simplify the computation, a multistage decomposition of optimization scheme is proposed and then a segmental recursive Riccati-like equation is derived. Moreover, in the case of time-invariant fuzzy systems, we show that the optimal controller can be obtained by just solving discrete-time algebraic Riccati-like equations. Grounding on this, several fascinating characteristics of the resultant closed-loop fuzzy system can be elicited easily. The stability of the closed-loop fuzzy system can be ensured by the designed optimal fuzzy controller. The optimal closed-loop fuzzy system can not only be guaranteed to be exponentially stable, but also be stabilized to any desired degree. Also, the total energy of system output is absolutely finite. Moreover, the resultant closed-loop fuzzy system possesses an infinite gain margin; that is, its stability is guaranteed no matter how large the feedback gain becomes. An example is given to illustrate the proposed optimal fuzzy controller design approach and to demonstrate the proved stability properties.
Lin, C.T., Duh, F.B. & Liu, D.J. 2002, 'A neural fuzzy network for word information processing', Fuzzy Sets and Systems, vol. 127, no. 1, pp. 37-48.
View/Download from: Publisher's site
A neural fuzzy system learning with fuzzy training data is proposed in this study. The system is able to process and learn numerical information as well as word information. At first, we propose a basic structure of five-layered neural network for the connectionist realization of a fuzzy inference system. The connectionist structure can house fuzzy logic rules and membership functions for fuzzy inference. The inputs, outputs, and weights of the proposed network can be fuzzy numbers of any shape. Also they can be hybrid of fuzzy numbers and numerical numbers through the use of fuzzy singletons. Based on interval arithmetics, a fuzzy supervised learning algorithm is developed for the proposed system. It extends the normal supervised learning techniques to the learning problems where only word teaching signals are available. The fuzzy supervised learning scheme can train the proposed system with desired fuzzy input-output pairs. An experimental system is constructed to illustrate the performance and applicability of the proposed scheme. &copy; 2002 Elsevier Science B.V. All rights reserved.
Liu, S.H. & Lin, C.T. 2001, 'A model-based fuzzy logic controller with Kalman filtering for tracking mean arterial pressure', IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans., vol. 31, no. 6, pp. 676-686.
View/Download from: Publisher's site
This paper proposes a new noninvasive measurement method for tracking the tendency of mean arterial pressure (MAP) in the radial artery. The designed system consists of a tonometer, a microsyringe device, and a model-based fuzzy logic controller. The modified flexible diaphragm tonometer is to detect the continuous blood pressure waveform and vessel volume pulse. A precise mathematical model describing the interaction between the tonometer and artery is derived. To reach accurate measurement without distortion, a model-based fuzzy logic control system is designed to compensate the change of MAP by applying a counter pressure on the tonometer chamber through the microsyringe device. The proposed control system consists of a linear predictor, a Kalman filter, and a synthetic fuzzy logic controller (SFLC). The linear predictor is to estimate the MAPs changing tendency based on the identified arterial pressure-volume model and then to beat-to-beat adjust the function of SFLC. The Kalman filter is to reduce the physiologic and measurement disturbance of the vessel volume oscillation amplitude (VOA). The SFLC is composed of three parallel subcontrollers, each of which is a simple fuzzy logic controller, for processing the three changing states of the MAP: ascending, descending, and stabilizing states, respectively. The design of the fuzzy rules in each subcontroller is based on the oscillometric principle saying that the arterial vessel has the maximum compliance when the detected vessel volume pulse reaches its maximum amplitude. Simulation results show that, for the real physiologic MAP with changing rates up to 20 or -20 mm-Hg/minute, the model-based SFLC can beat-to-beat adjust the tonometer's chamber pressure to follow the tendency of MAP accurately.
Liu, D.J. & Lin, C.T. 2001, 'Fundamental frequency estimation based on the joint time-frequency analysis of harmonic spectral structure', IEEE Transactions on Speech and Audio Processing, vol. 9, no. 6, pp. 609-621.
View/Download from: Publisher's site
In this paper, we propose a new scheme to analyze the spectral structure of speech signals for fundamental frequency estimation. First, we propose a pitch measure to detect the harmonic characteristics of voiced sounds on the spectrum of a speech signal. This measure utilizes the properties that there are distinct impulses located at the positions of fundamental frequency and its harmonics, and the energy of voiced sound is dominated by the energy of these distinct harmonic impulses. The spectrum can be obtained by the fast Fourier transform (FFT); however, it may be destroyed when the speech is interfered with by additive noise. To enhance the robustness of the proposed scheme in noisy environments, we apply the joint time-frequency analysis (JTFA) technique to obtain the adaptive representation of the spectrum of speech signals. The adaptive representation can accurately extract important harmonic structure of noisy speech signals at the expense of high computation cost. To solve this problem, we further propose a fast adaptive representation (FAR) algorithm, which reduces the computation complexity of the original algorithm by 50%. The performance of the proposed fundamental-frequency estimation scheme is evaluated on a large database with or without additive noise. The performance is compared to that of other approaches on the same database. The experimental results show that the proposed scheme performs well on clean speech and is robust in noisy environments.
Lin, C.T., Chung, I.F. & Huang, S.Y. 2001, 'Improvement of machining accuracy by fuzzy logic at corner parts for wire-EDM', Fuzzy Sets and Systems, vol. 122, no. 3, pp. 499-511.
View/Download from: Publisher's site
Wire electrical discharge machining (wire-EDM) has always occupied an important position in some production fields, due to its capability of machining hard materials and intricate shapes. However, the machining accuracy, especially at corner parts, may be destroyed because of some phenomena such as wire deflection and vibration, etc. The purpose of this paper is to develop a control strategy based on fuzzy logic so that the machining accuracy at corner parts for wire-EDM can be improved. The fuzzy rules based on the wire-EDM's physical characteristics, experimental data, and operator's experience are constructed, so that the reduced percentage of sparking force can be determined by a multi-variables fuzzy logic controller. The objective of the total control is to improve the machining accuracy at corner parts, but still keep the cutting feedrate at fair values. As a result of experiments, machining errors of corner parts, especially in rough-cutting, can be reduced to less than 50% of those in normal machining, while the machining process time increases not more than 10% of the normal value. &copy; 2001 Elsevier Science B.V.
Juang, C.F. & Lin, C.T. 2001, 'Noisy speech processing by recurrently adaptive fuzzy filters', IEEE Transactions on Fuzzy Systems, vol. 9, no. 1, pp. 139-152.
View/Download from: Publisher's site
Two noisy speech processing problems-speech enhancement and noisy speech recognition-are dealt with in this paper. The technique we focus on is by using the filtering approach; a novel filter, the recurrently adaptive fuzzy filter (RAFF), is proposed and applied to these two problems. The speech enhancement is based on adaptive noise cancellation with two microphones, where the RAFF is used to eliminate the noise corrupting the desired speech signal in the primary channel. As to the noisy speech recognition, the RAFF is used to filter the noise in the feature domain of speech signals. The RAFF is inherently a recurrent multilayered connectionist network for realizing the basic elements and functions of dynamic fuzzy inference, and may be considered to be constructed from a series of dynamic fuzzy rules. As compared to other existing nonlinear filters, three major advantages of the RAFF are observed: 1) a priori knowledge can be incorporated into the RAFF, which makes the fusion of numerical data and linguistic information possible; 2) owing to the dynamic property of the RAFF, the exact lagged order of the input variables need not be known in advance; 3) no predetermination, like the number of hidden nodes, must be given since the RAFF can find its optimal structure and parameters automatically. Several examples on adaptive noise cancellation and noisy speech recognition problems using the RAFF are illustrated to demonstrate the performance of the RAFF.
Wu, G.D. & Lin, C.T. 2001, 'A recurrent neural fuzzy network for word boundary detection in variable noise-level environments', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 31, no. 1, pp. 84-97.
View/Download from: Publisher's site
This paper discusses the problem of automatic word boundary detection in the presence of variable-level background noise. Commonly used robust word boundary detection algorithms always assume that the background noise level is fixed. In fact, the background noise level may vary during the procedure of recording. This is the major reason that most robust word boundary detection algorithms cannot work well in the condition of variable background noise level. In order to solve this problem, we first propose a refined time-frequency (RTF) parameter for extracting both the time and frequency features of noisy speech signals. The RTF parameter extends the (time-frequency) TF parameter proposed by Junqua et al. from single band to multiband spectrum analysis, where the frequency bands help to make the distinction between speech signal and noise clear. The RTF parameter can extract useful frequency information. Based on this RTF parameter, we further propose a new word boundary detection algorithm by using a recurrent self-organizing neural fuzzy inference network (RSONFIN). Since RSONFIN can process the temporal relations, the proposed RTF-based RSONFIN algorithm can find the variation of the background noise level and detect correct word boundaries in the condition of variable background noise level. As compared to normal neural networks, the RSONFIN can always find itself an economic network size with high-learning speed. Due to the self-learning ability of RSONFIN, this RTF-based RSONFIN algorithm avoids the need for empirically determining ambiguous decision rules in normal word boundary detection algorithms. Experimental results show that this new algorithm achieves higher recognition rate than the TF-based algorithm which has been shown to outperform several commonly used word boundary detection algorithms by about 12% in variable background noise level condition. It also reduces the recognition error rate due to endpoint detection to about 23%, compared to an ave...
Nein, H.W. & Lin, C.T. 2001, 'Incorporating error shaping technique into LSF vector quantization', IEEE Transactions on Speech and Audio Processing, vol. 9, no. 2, pp. 73-86.
View/Download from: Publisher's site
This paper presents an error shaping technique for line spectrum frequency (LSF) vector quantization. The error shaping technique based on the weighted logarithm spectral distortion (WLSD) measure can be used for shaping the spectral distortion distribution of quantization error into any different curve depending on what kind of weighting function is used. However, the high computational complexity of the WLSD measure deters this error shaping technique from practical use. To solve this problem, we approximate the WLSD measure by the quadratically weighted measure or the weighted mean squared error (WMSE) measure and propose an optimal error shaping technique of LSF vector quantization. In this proposed error shaping technique, the optimal WMSE weights (i.e., the optimal weights of LSF parameters) are determined based on the theoretical analysis of the WLSD measure. Three experiments are performed to check the performance of the proposed error shaping technique. One experiment is set up by incorporating human perception into the LSF quantization and another is set up by emphasizing the human-sensitivity frequency band in lower frequency bandwidth 0-3 kHz. In the third experiment, we apply the proposed error shaping technique to the LSF quantization of a CELP coder to test how it affects the overall speech quality in an actual speech coding algorithm.
Wang, C.H., Liu, H.L. & Lin, C.T. 2001, 'Dynamic optimal learning rates of a certain class of fuzzy neural networks and its applications with genetic algorithm', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 31, no. 3, pp. 467-475.
View/Download from: Publisher's site
The stability analysis of the learning rate for a two-layer neural network (NN) is discussed first by minimizing the total squared error between the actual and desired outputs for a set of training vectors. The stable and optimal learning rate, in the sense of maximum error reduction, for each iteration in the training (back propagation) process can therefore be found for this two-layer NN. It has also been proven in this paper that the dynamic stable learning rate for this two-layer NN must be greater than zero. Thus it is guaranteed that the maximum error reduction can be achieved by choosing the optimal learning rate for the next training iteration. A dynamic fuzzy neural network (FNN) that consists of the fuzzy linguistic process as the premise part and the two-layer NN as the consequence part is then illustrated as an immediate application of our approach. Each part of this dynamic FNN has its own learning rate for training purpose. A genetic algorithm is designed to allow a more efficient tuning process of the two learning rates of the FNN. The objective of the genetic algorithm is to reduce the searching time by searching for only one learning rate, which is the learning rate of the premise part, in the FNN. The dynamic optimal learning rates of the two-layer NN can be found directly using our innovative approach. Several examples are fully illustrated and excellent results are obtained for the model car backing up problem and the identification of nonlinear first order and second order systems.
Wu, S.J. & Lin, C.T. 2001, 'Optimal fuzzy tracking controller design for discrete-time fuzzy systems', Zidonghua Xuebao/Acta Automatica Sinica, vol. 27, no. 4, pp. 477-494.
In this paper, we propose a systematic and theoretically sound way to design a global optimal fuzzy tracking controller for discrete-time fuzzy systems with the aim of solving the discrete-time quadratic tracking problems with moving or model-following targets under finite or infinite horizon (time). A linear-like dynamical system representation of discrete-time fuzzy system is proposed to mature the theoretical design scheme of the discrete-time optimal fuzzy tracking controller which can achieve global minimum effect. A multistage decomposition of optimization scheme is proposed to simplify the computation, and then a segmental recursive Riccati-like equation and a difference equation in tracing the variation of the target are derived. Moreover, in the case of time-invariant fuzzy tracking systems, we show that the optimal tracking controller can be obtained by just solving discrete-time algebraic Riccati-like equations and algebraic matrix equations. An example is given to illustrate the proposed opti mal fuzzy tracker design scheme.
Bai, M.R., Hsiao, I., Tsai, H. & Lin, C. 2000, 'Development of an on-line diagnosis system for rotor vibration via model-based intelligent inference', The Journal of the Acoustical Society of America, vol. 107, no. 1, pp. 315-323.
An on-line fault detection and isolation technique is proposed for the diagnosis of rotating machinery. The architecture of the system consists of a feature generation module and a fault inference module. Lateral vibration data are used for calculating the system features. Both continuous-time and discrete-time parameter estimation algorithms are employed for generating the features. A neural fuzzy network is exploited for intelligent inference of faults based on the extracted features. The proposed method is implemented on a digital signal processor. Experiments carried out for a rotor kit and a centrifugal fan indicate the potential of the proposed techniques in predictive maintenance.
Chung, I.F., Lin, C.J. & Lin, C.T. 2000, 'A GA-based fuzzy adaptive learning control network', Fuzzy Sets and Systems, vol. 112, no. 1, pp. 65-84.
This paper addresses the structure and an associated learning algorithm of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed fuzzy adaptive learning control network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. A structure/parameter learning algorithm, called FALCON-GA, is proposed for constructing the FALCON automatically. The FALCON-GA is a three-phase hybrid learning algorithm. In the first phase, the fuzzy ART algorithm is used to do fuzzy clustering in the input/output spaces according to the supervised training data. In the second phase, the genetic algorithm (GA) is used to find proper fuzzy logic rules by associating input clusters and output clusters. Finally, in the third phase, the backpropagation algorithm is used for tuning input/output membership functions. Hence, the FALCON-GA combines the backpropagation algorithm for parameter learning and both the fuzzy ART and GAs for structure learning. It can partition the input/output spaces, tune membership functions and find proper fuzzy logic rules automatically. The proposed FALCON has two important features. First, it reduces the combinatorial demands placed by the standard methods for adaptive linearization of a system. Second, the FALCON is a highly autonomous system. In its learning scheme, only the training data need to be provided from the outside world. The users need not give the initial fuzzy partitions, membership functions and fuzzy logic rules. Computer simulations have been conducted to illustrate the performance and applicability of the proposed system. &copy; 2000 Elsevier Science B.V. All rights reserved.
Wang, Y.J. & Lin, C.T. 2000, 'Recurrent learning algorithms for designing optimal controllers of continuous systems', IEEE Transactions on Systems, Man, and Cybernetics Part A:Systems and Humans., vol. 30, no. 5, pp. 580-588.
View/Download from: Publisher's site
This paper proposes the recurrent learning algorithm for designing the controllers of continuous dynamical systems in the optimal control problems. The designed controllers are in the form of unfolded recurrent neural networks embedded with physical laws coming from the classical control techniques. The proposed learning algorithm is characterized by its double-forward-recurrent-loops structure for solving both the temporal recurrent and the structure recurrent problems. The first problem is resulted from the nature of general optimal control problems, where the objective functions are often related to (evaluated at) some specific (instead of all) time steps or system states only, causing missing learning signals at some time steps or system states. The second problem is due to the high-order discretization of the continuous systems by the Runge-Kutta method that we perform to increase the control accuracy. This discretization transforms the system into several identical subnetworks interconnected together, like a recurrent neural network expanded in the time axis. Two recurrent learning algorithms with different convergence properties are derived; the first- and second-order learning algorithms. The computations of both algorithms are local and performed efficiently as network signal propagation. We also propose two new nonlinear controller structures for two specific control problems: 1) two-dimensional (2-D) guidance problem and 2) optimal PI control problem. Under the training of the proposed recurrent learning algorithms, these two controllers can be easily tuned to be suboptimal for given objective functions. Extensive computer simulations have shown the optimization and generalization abilities of the controllers designed by the proposed learning scheme.
Lin, C.T., Juang, C.F. & Li, C.P. 2000, 'Water bath temperature control with a neural fuzzy inference network', Fuzzy Sets and Systems, vol. 111, no. 2, pp. 285-306.
View/Download from: Publisher's site
Although multilayered backpropagation neural networks (BPNN) have demonstrated high potential in the nonconventional branch of adaptive control, its long training time usually discourages their applications in industry. Moreover, when they are trained on-line to adapt to plant variations, the overtuned phenomenon usually occurs. To overcome the weakness of the BPNN, we propose a neural fuzzy inference network (NFIN) in this paper suitable for adaptive control of practical plant systems in general, and for adaptive temperature control of a water bath system in particular. The NFIN is inherently a modified TSK (Takagi-Sugeno-Kang)-type fuzzy rule-based model possessing neural network's learning ability. In contrast to the general adaptive neural fuzzy networks, where the rules should be decided in advance before parameter learning is performed, there are no rules initially in the NFIN. The rules in the NFIN are created and adapted as on-line learning proceeds via simultaneous structure and parameter identification. The NFIN has been applied to a water bath temperature control system. As compared to the BPNN under the same training procedure, the control results show that not only can the NFIN greatly reduce the training time and avoid the overtuned phenomenon, but the NFIN also has perfect regulation ability.
Juang, C.F., Lin, J.Y. & Lin, C.T. 2000, 'Genetic reinforcement learning through symbiotic evolution for fuzzy controller design', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 30, no. 2, pp. 290-302.
View/Download from: Publisher's site
An efficient genetic reinforcement learning algorithm for designing fuzzy controllers is proposed in this paper. The genetic algorithm (GA) adopted in this paper is based upon symbiotic evolution which, when applied to fuzzy controller design, complements the local mapping property of a fuzzy rule. Using this Symbiotic-Evolution-based Fuzzy Controller (SEFC) design method, the number of control trials, as well as consumed CPU time, are considerably reduced when compared to traditional GA-based fuzzy controller design methods and other types of genetic reinforcement learning schemes. Moreover, unlike traditional fuzzy controllers, which partition the input space into a grid, SEFC partitions the input space in a flexible way, thus creating fewer fuzzy rules. In SEFC, different types of fuzzy rules whose consequent parts are singletons, fuzzy sets, or linear equations (TSK-type fuzzy rules) are allowed. Further, the free parameters (e.g., centers and widths of membership functions) and fuzzy rules are all tuned automatically. For the TSK-type fuzzy rule especially, which put the proposed learning algorithm in use, only the significant input variables are selected to participate in the consequent of a rule. The proposed SEFC design method has been applied to different simulated control problems, including the cart-pole balancing system, a magnetic levitation system, and a water bath temperature control system. The proposed SEFC has been verified to be efficient and superior from these control problems, and from comparisons with some traditional GA-based fuzzy systems.
Lin, C.T., Chung, I.F. & Sheu, L.K.M. 2000, 'A neural fuzzy system for image motion estimation', Fuzzy Sets and Systems, vol. 114, no. 2, pp. 281-304.
Many methods for computing optical flow (image motion vector) have been proposed while others continue to appear. Block-matching methods are widely used because of their simplicity and easy implementation. The motion vector is uniquely defined, in block-matching methods, by the best fit of a small reference subblock from a previous image frame in a larger, search region from the present image frame. Hence, this method is very sensitive to the real environments (involving occlusion, specularity, shadowing, transparency, etc.). In this paper, a neural fuzzy system with robust characteristics and learning ability is incorporated with the block-matching method to make a system adaptive for different circumstances. In the neural fuzzy motion estimation system, each subblock in the search region is assigned a similarity membership contributing different degrees to the motion vector. This system is more reliable, robust, and accurate in motion estimation than many other methods including Horn and Schunck's optical flow, fuzzy logic motion estimator (FME), best block matching, NR, and fast block matching. Since fast block-matching algorithms can be used to reduce search time, a three-step fast search method is employed to find the motion vector in our system. However, the candidate motion vector is often trapped by the local minimum, which makes the motion vector undesirable. An improved three-step fast search method is tested to reduce the effect from local minimum and some comparisons about fast search algorithms are made. In addition, a Quarter Compensation Algorithm for compensating the interframe image to tackle the problem that the motion vector is not an integer but rather a floating point is proposed. Since our system can give the accurate motion vector, we may use the motion information in many different applications such as motion compensation, CCD camera auto-focusing or zooming, moving object extraction, etc. Two application examples will be illustrated in th...
Liang, S.F., Su, A.W.Y. & Lin, C.T. 2000, 'Model-based synthesis of plucked string instruments by using a class of scattering recurrent networks', IEEE Transactions on Neural Networks, vol. 11, no. 1, pp. 171-185.
View/Download from: Publisher's site
A physical modeling method for electronic music synthesis of plucked-string tones by using recurrent networks is proposed. A scattering recurrent network (SRN) which is used to analyze string dynamics is built based on the physics of acoustic strings. The measured vibration of a plucked string is employed as the training data for the supervised learning of the SRN. After the network is well trained, it can be regarded as the virtual model for the measured string and used to generate tones which can be very close to those generated by its acoustic counterpart. The `virtual string' corresponding to the SRN can respond to different `plucks' just like a real string, which is impossible using traditional synthesis techniques such as frequency modulation and wavetable. The simulation of modeling a cello `A'-string demonstrates some encouraging results of the new music synthesis technique. Some aspects of modeling and synthesis procedures are also discussed.
Wu, S.J. & Lin, C.T. 2000, 'Optimal fuzzy controller design: Local concept approach', IEEE Transactions on Fuzzy Systems, vol. 8, no. 2, pp. 171-185.
View/Download from: Publisher's site
In this paper, we present a global optimal and stable fuzzy controller design method for both continuous- and discrete-time fuzzy systems under both finite and infinite horizons. First, a sufficient condition is proposed which indicates that the global optimal effect can be achieved by the fuzzily combined local optimal controllers. Based on this sufficient condition, we derive a local concept approach to designing the optimal fuzzy controller by applying traditional linear optimal control theory. The stability of the entire closed-loop continuous fuzzy system can be ensured by the designed optimal fuzzy controller. The optimal feedback continuous fuzzy system can not only be guaranteed to be exponentially stable, but also be stabilized to any desired degree. Also, the total energy of system output is absolutely finite. Moreover, the resultant feedback continuous fuzzy system possesses an infinite gain margin; that is, its stability is guaranteed no matter how large the feedback gain becomes. Two examples are given to illustrate the proposed optimal fuzzy controller design approach and to demonstrate the proved stability properties.
Lin, C.T., Nein, H.W. & Hwu, J.Y. 2000, 'GA-based noisy speech recognition using two-dimensional cepstrum', IEEE Transactions on Speech and Audio Processing, vol. 8, no. 6, pp. 664-674.
View/Download from: Publisher's site
Abstract-Among various kinds of speech features, the two-dimensional (2-D) cepstrum (TDC) is a special one, which can simultaneously represent several types of information contained in the speech waveform: static and dynamic features, as well as global and fine frequency structures. Analysis results show that the coefficients located at lower indexes portion of the TDC matrix seem to be more significant than others. Hence, to represent an utterance only some TDC coefficients need to be selected to form a feature vector instead of the sequence of feature vectors. It has the advantages of simple computation and less storage space. However, our experiments show that the selection of TDC coefficients is quite sensitive to background noise. In order to solve this problem, we propose the GA-based M_TDC (modified TDC) method in this paper to improve the representativeness and robustness of the selected TDC coefficients in noisy environments. The M_TDC differs from the standard TDC by the use of filters to remove the noise components. Furthermore, in the GA-based M_TDC method, we apply the genetic algorithms (GAs) to find the robust coefficients in the M_TDC matrix. From the experiments with five noise types, we find that the GA-based M_TDC method has better recognition results than the original TDC approach in noisy environments. &copy; 2000 IEEE Publisher Item Identifier S 1063-6676(00)09268-3.
Wu, G.D. & Lin, C.T. 2000, 'Word boundary detection with mel-scale frequency bank in noisy environment', IEEE Transactions on Speech and Audio Processing, vol. 8, no. 5, pp. 541-553.
View/Download from: Publisher's site
This paper addresses the problem of automatic word boundary detection in the presence of noise. We first propose an adaptive time-frequency (ATF) parameter for extracting both the time and frequency features of noisy speech signals. The ATF parameter extends the TF parameter proposed by Junqua et al. from single band to multiband spectrum analysis, where the frequency bands help to make the distinction of speech and noise signals clear. The ATF parameter can extract useful frequency information by adoptively choosing proper bands of the inel-scale frequency bank. The ATF parameter increased the recognition rate by about 3% of a TF-based robust algorithm which has been shown to outperform several commonly used algorithms for word boundary detection in the presence of noise. The ATF parameter also reduced the recognition error rate due to endpoint detection to about 20%. Based on the ATF parameter, we further propose a new word boundary detection algorithm by using a neural fuzzy network (called SONFIN) for identifying islands of word signals in noisy environment. Due to the self-learning ability of SONFIN, the proposed algorithm avoids the need of empirically determining thresholds and ambiguous rules in normal word boundary detection algorithms. As compared to normal neural networks, the SONFIN can always find itself an economic network size, in high learning speed. Our results also showed that the SONFIN's performance is not significantly affected by the size of training set. The ATF-based SONFIN achieved higher recognition rate than the TF-based robust algorithm by about 5%. It also reduced the recognition error rate due to endpoint detection to about 10%, compared to an average of approximately 30% obtained with the TF-based robust algorithm, and 50% obtained with the modified version of the Lamel et al. algorithm. &copy; 2000 IEEE.
Wu, S.J. & Lin, C.T. 2000, 'Optimal fuzzy controller design in continuous fuzzy system: global concept approach', IEEE Transactions on Fuzzy Systems, vol. 8, no. 6, pp. 713-729.
View/Download from: Publisher's site
In this paper, we propose a systematic and theoretically sound way to design a global optimal fuzzy controller to control and stabilize a continuous fuzzy system with free- or fixed-end point under finite or infinite horizon (time). A linear-like global system representation of continuous fuzzy system is first proposed by viewing a continuous fuzzy system in global concept and unifying the individual matrices into synthetical matrices. Based on this, the optimal control law which can achieve global minimum effect is developed theoretically. The nonlinear segmental two-point boundary-value problem (TPBVP) is derived for the finite-horizon problem and a forward Riccati-like differential equation (DE) for the infinite-horizon problem. To further simplify the computation, a segmental Riccati-like DE is derived in solving the finite- or infinite-horizon issues. Moreover, in the case of time-invariant fuzzy systems, we show that the optimal controller can be obtained by just solving algebraic Riccati-like equations. Grounding on this, several fascinating characteristics of the resultant closed-loop fuzzy system can be elicited easily. The stability of the closed-loop fuzzy system can be ensured by the designed optimal fuzzy controller. The optimal closed-loop fuzzy system cannot only be guaranteed to be exponentially stable, but also be stabilized to any desired degree. Also, the total energy of system output is absolutely finite. Moreover, the resultant closed-loop fuzzy system possesses an infinite gain margin; that is, its stability is guaranteed no matter how large the feedback gain becomes. An example is given to illustrate the proposed optimal fuzzy controller design approach and to demonstrate the proved stability properties.
Shieh, C.S. & Lin, C.T. 2000, 'Direction of arrival estimation based on phase differences using neural fuzzy network', IEEE Transactions on Antennas and Propagation, vol. 48, no. 7, pp. 1115-1124.
View/Download from: Publisher's site
A new high-resolution direction of arrival (DOA) estimation technique using a neural fuzzy network based on phase difference (PD) is proposed in this paper. The conventional DOA estimation method such as MUSIC and MLE, are computationally intensive and difficult to implement in real time. To attach these problems, neural networks have become popular for DOA estimation in recent years. However, the normal neural networks such as multilayer perceptron (MLP) and radial basis function network (RBFN) usually produce the extra problems of low convergence speed and/or large network size (i.e., the number of network parameters is large). Also, the way to decide the network structure is heuristic. To overcome these defects and take use of neural learning ability, a powerful self-constructing neural fuzzy inference network (SONFIN) is used to develop a new DOA estimation algorithm in this paper. By feeding the PD's of received radar-array signals, the trained SONFIN can give high-resolution DOA estimation. The proposed scheme is thus called PD-SONFIN. This new algorithm avoids the need of empirically determining the network size and parameters in normal neural networks due to the powerful on-line structure and parameter learning ability of SONFIN. The PD-SONFIN can always find itself an economical network size in fast learning process. Our simulation results show that the performance of the new algorithm is superior to the RBFN in terms of convergence accuracy, estimation accuracy, sensitivity to noise, and network size.
Lin, C.T. & Jou, C.P. 2000, 'GA-based fuzzy reinforcement learning for control of a magnetic bearing system', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 30, no. 2, pp. 276-289.
View/Download from: Publisher's site
This paper proposes a TD (temporal difference) and GA (genetic algorithm)-based reinforcement (TDGAR) learning method and applies it to the control of a real magnetic bearing system. The TDGAR learning scheme is a new hybrid GA, which integrates the TD prediction method and the GA to perform the reinforcement learning task. The TDGAR learning system is composed of two integrated feedforward networks. One neural network acts as a critic network to guide the learning of the other network (the action network) which determines the outputs (actions) of the TDGAR learning system. The action network can be a normal neural network or a neural fuzzy network. Using the TD prediction method, the critic network can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the action network. The action network uses the GA to adapt itself according to the internal reinforcement signal. The key concept of the TDGAR learning scheme is to formulate the internal reinforcement signal as the fitness function for the GA such that the GA can evaluate the candidate solutions (chromosomes) regularly, even during periods without external feedback from the environment. This enables the GA to proceed to new generations regularly without waiting for the arrival of the external reinforcement signal. This can usually accelerate the GA learning since a reinforcement signal may only be available at a time long after a sequence of actions has occurred in the reinforcement learning problem. The proposed TDGAR learning system has been used to control an active magnetic bearing (AMB) system in practice. A systematic design procedure is developed to achieve successful integration of all the subsystems including magnetic suspension, mechanical structure, and controller training. The results show that the TDGAR learning scheme can successfully find a neural controller or a neural fuzzy controller for a self-designed magnetic bearing system.
Lin, C.T., Lee, Y.C. & Pu, H.C. 2000, 'Satellite sensor image classification using cascaded architecture of neural fuzzy network', IEEE Transactions on Geoscience and Remote Sensing, vol. 38, no. 2 II, pp. 1033-1043.
View/Download from: Publisher's site
Satellite sensor images usually contain many complex factors and mixed pixels, so a high classification accuracy is not easy to attain. Especially, for a nonhomogeneous region, gray values of satellite sensor images vary greatly and thus, direct statistic gray values fail to do the categorization task correctly. The goal of this paper is to develop a cascaded architecture of neural fuzzy networks with feature mapping (CNFM) to help the clustering of satellite sensor images. In the CNFM, a Kohonen's self-organizing feature map (SOFM) is used as a preprocessing layer for the reduction of feature domain, which combines original multi-spectral gray values, structural measurements from co-occurrence matrices, and spectrum features from wavelet decomposition. In addition to the benefit of dimensional reduction of feature space, Kohonen's SOFM can remove some noisy areas and prevent the following training process from being overoriented to the training patterns. The condensed measurements are then forwarded into a neural fuzzy network, which performs supervised learning for pattern classification. The proposed cascaded approach is an appropriate technique for handling the classification problem in areas that exhibit large spatial variation and interclass heterogeneity (e.g., urban-rural infringing areas). The CNFM is a general and useful structure that can give us favorable results in terms of classification accuracy and learning speed. Experimental results indicate that our structure can retain high accuracy of classification (90% in average), while the training time is substantially reduced if our system is compared to the commonly used backpropagation network. The CNFM appears to be more reasonable and practical than the conventional implementation.
Hsiao, I.L., Kuo, C.H., Bai, M.S. & Lin, C.T. 1999, 'On-line fault diagnosis of rotor vibration by using signal-based feature generation and neural fuzzy inference', Journal of the Chinese Society of Mechanical Engineers, Transactions of the Chinese Institute of Engineers, Series C/Chung-Kuo Chi Hsueh Kung Ch'eng Hsuebo Pao, vol. 20, no. 4, pp. 345-352.
An on-line fault detection and isolation technique is proposed for the diagnosis of rotor vibration. The architecture of the systems mainly consists of feature generation and fault inference. A signal-based method is used for generating the features required by the subsequent neural fuzzy inference. In the signal-based approach, both lateral and axial vibration data are used for calculating signal features such as the average, the standard deviation, the maximum, and the harmonic multiples. A neural fuzzy network is exploited for intelligent inference of faults based on the extracted features. The proposed systems are implemented on the platform of a digital signal processor. Experiments carried out for a rotor kit, a centrifugal fan, and a centrifugal pump indicate the potential of the proposed techniques in predictive maintenance.
Lai, J.H. & Lin, C.T. 1999, 'Application of neural fuzzy network to pyrometer correction and temperature control in rapid thermal processing', IEEE Transactions on Fuzzy Systems, vol. 7, no. 2, pp. 160-175.
View/Download from: Publisher's site
Temperature measurement and control are two difficult problems in the rapid thermal processing (RTP) system. For many applications such as rapid thermal processing chemical vapor deposition (RTCVD) and rapid thermal oxidation (RTO), large changes in wafer emissivity can occur during film growing, leading to erroneous temperature measurements with a single wavelength pyrometer. The error in the inferred temperature will affect the temperature control of the RTP system. In order to correct the temperature reading of the pyrometer, a neural fuzzy network is used to predict the emissivity changes for the compensation of measured temperature. As for the temperature control, to overcome ill performance of the temperature tracking system due to the inaccuracy of the identified model, another neural fuzzy network is used in the RTP system for learning inverse control simultaneously. The key advantage of neural fuzzy approach over traditional ones lies on that the approach does not require a mathematical description of the system while performing pyrometer correction and temperature control. Simulation results show that the adopted neural fuzzy networks can not only correct the pyrometer reading accurately, but also be able to track a temperature trajectory very well.
Lin, C.T. & Chung, I.F. 1999, 'A reinforcement neuro-fuzzy combiner for multiobjective control', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 29, no. 6, pp. 726-744.
View/Download from: Publisher's site
This paper proposes a neuro-fuzzy combiner (NFC) with reinforcement learning capability for solving multiobjective control problems. The proposed NFC can combine n existing low-level controllers in a hierarchical way to form a multiobjective fuzzy controller. It is assumed that each low-level (fuzzy or nonfuzzy) controller has been well designed to serve a particular objective. The role of the NFC is to fuse the n actions decided by the n low-level controllers and determine a proper action acting on the environment (plant) at each time step. Hence, the NFC can combine low-level controllers and achieve multiple objectives (goals) at once. The NFC acts like a switch that chooses a proper action from the actions of low-level controllers according to the feedback information from the environment. In fact, the NFC is a soft switch; it allows more than one low-level actions to be active with different degrees through fuzzy combination at each time step. An NFC can be designed by the trial-and-error approach if enough a priori knowledge is available, or it can be obtained by supervised learning if precise input/output training data are available. In the more practical cases when there is no instructive teaching information available, the NFC can learn by itself using the proposed reinforcement learning scheme. Adopted with reinforcement learning capability, the NFC can learn to achieve desired multiobjectives simultaneously through the rough reinforcement feedback from the environment, which contains only critic information such as "success (good)" or "failure (bad)" for each desired objective. Computer simulations have been conducted to illustrate the performance and applicability of the proposed architecture and learning scheme. &copy; 1999 IEEE.
Lin, C.T., Nein, H.W. & Lin, W.C. 1999, 'A space-time delay neural network for motion recognition and its application to lipreading.', International journal of neural systems, vol. 9, no. 4, pp. 311-334.
Motion recognition has received increasing attention in recent years owing to heightened demand for computer vision in many domains, including the surveillance system, multimodal human computer interface, and traffic control system. Most conventional approaches classify the motion recognition task into partial feature extraction and time-domain recognition subtasks. However, the information of motion resides in the space-time domain instead of the time domain or space domain independently, implying that fusing the feature extraction and classification in the space and time domains into a single framework is preferred. Based on this notion, this work presents a novel Space-Time Delay Neural Network (STDNN) capable of handling the space-time dynamic information for motion recognition. The STDNN is unified structure, in which the low-level spatiotemporal feature extraction and high-level space-time-domain recognition are fused. The proposed network possesses the spatiotemporal shift-invariant recognition ability that is inherited from the time delay neural network (TDNN) and space displacement neural network (SDNN), where TDNN and SDNN are good at temporal and spatial shift-invariant recognition, respectively. In contrast to multilayer perceptron (MLP), TDNN, and SDNN, STDNN is constructed by vector-type nodes and matrix-type links such that the spatiotemporal information can be accurately represented in a neural network. Also evaluated herein is the performance of the proposed STDNN via two experiments. The moving Arabic numerals (MAN) experiment simulates the object's free movement in the space-time domain on image sequences. According to these results, STDNN possesses a good generalization ability with respect to the spatiotemporal shift-invariant recognition. In the lipreading experiment, STDNN recognizes the lip motions based on the inputs of real image sequences. This observation confirms that STDNN yields a better performance than the existing TDNN-based syst...
Lin, C.T., Nein, H.W. & Lin, W.F. 1999, 'Speaker adaptation of fuzzy-perceptron-based speech recognition', International Journal of Uncertainty, Fuzziness and Knowlege-Based Systems, vol. 7, no. 1, pp. 1-30.
In this paper, we propose a speech recognition algorithm which utilizes hidden Markov models (HMM) and Viterbi algorithm for segmenting the input speech sequence, such that the variable-dimensional speech signal is converted into a fixed-dimensional speech signal, called TN vector. We then use the fuzzy perceptron to generate hyperplanes which separate patterns of each class from the others. The proposed speech recognition algorithm is easy for speaker adaptation when the idea of "supporting pattern" is used. The supporting patterns are those patterns closest to the hyperplane. When a recognition error occurs, we include all the TN vectors of the input speech sequence with respect to the segmentations of all HMM models as the supporting patterns. The supporting patterns are then used by the fuzzy perceptron to tune the hyperplane that can cause correct recognition, and also tune the hyperplane that resulted in wrong recognition. Since only two hyperplanes need to be tuned for a recognition error, the proposed adaptation scheme is time-economic and suitable for on-line adaptation. Although the adaptation scheme cannot ensure to correct the wrong recognition right after adaptation, the hyperplanes are tuned in the direction for correct recognition iteratively and the speed of adaptation can be adjusted by a "belief" parameter set by the user. Several examples are used to show the performance of the proposed speech recognition algorithm and the speaker adaptation scheme.
Lin, C.T. & Jou, C.P. 1999, 'Controlling chaos by GA-based reinforcement learning neural network', IEEE Transactions on Neural Networks, vol. 10, no. 4, pp. 846-859.
View/Download from: Publisher's site
This paper proposes a TD (temporal difference) and GA (genetic algorithm) based reinforcement (TDGAR) neural learning scheme for controlling chaotic dynamical systems based on the technique of small perturbations. The TDGAR learning scheme is a new hybrid GA, which integrates the TD prediction method and the GA to fulfill the reinforcement learning task. Structurely, the TDGAR learning system is composed of two integrated feedforward networks. One neural network acts as a critic network for helping the learning of the other network, the action network, which determines the outputs (actions) of the TDGAR learning system. Using the TD prediction method, the critic network can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the action network. The action network uses the GA to adapt itself according to the internal reinforcement signal. This can usually accelerate the GA learning since an external reinforcement signal may only be available at a time long after a sequence of actions have occurred in the reinforcement learning problems. By defining a simple external reinforcement signal, the TDGAR learning system can learn to produce a series of small perturbations to convert chaotic oscillations of a chaotic system into desired regular ones with a periodic behavior. The proposed method is an adaptive search for the optimum control technique. Computer simulations on controlling two chaotic systems, i.e., the Henon map and the logistic map, have been conducted to illustrate the performance of the proposed method.
Lin, C.T., Wu, G.D. & Hsiao, S.C. 1999, 'New techniques on deformed image motion estimation and compensation', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 29, no. 6, pp. 846-859.
View/Download from: Publisher's site
In this paper, new techniques for deformed image motion estimation and compensation using variable-size block-matching are proposed, which can be applied to an image sequence compression system or a moving object recognition system. The motion estimation and compensation techniques have been successfully applied in the area of image sequence coding. Many research papers on improving the performance of these techniques have been published; many directions are proposed, which can all lead to better performance than the conventional techniques. Among them, both generalized block-matching and variable-size block-matching are successfully applied in reducing the data rate of compensation error and motion information, respectively. These two algorithms have their merits, but suffer from their drawbacks. Moreover, reducing the data rate in compensation error is sometimes increasing the data rate in motion information, or vice versa. Based on these two algorithms, we propose and examine several algorithms which are effective in reducing the data rate. We then incorporate these algorithms into a system, in which they work together to overcome the disadvantages of individual and keep their merits at the same time. The proposed system can optimally balance the amount of data rate in two aspects (i.e., compensation error and motion information). Experimental results show that the proposed system outweighs the conventional techniques. Since we propose a recovery operation which tries to recover the incorrect motion vectors from the global motion, this proposed system can also be applied for the moving object recognition in image sequence. &copy; 1999 IEEE.
Lin, C.T., Juang, C.F. & Huang, J.C. 1999, 'Temperature control of rapid thermal processing system using adaptive fuzzy network', Fuzzy Sets and Systems, vol. 103, no. 1, pp. 49-65.
Temperature control of a rapid thermal processing (RTP) system using a proposed self-constructing adaptive fuzzy inference network (SCAFIN) is presented in this paper. First, the physical modeling of a RTP system is done. An integrated model is given for the components that make up a RTP system. These components are the lamp power dynamics, ray-tracing model, and the wafer thermal dynamic model. The models for the components are integrated in a numerical code to give a computer simulation of the complete RTP system. The simulation can be used to investigate the interaction of the furnace, lamp contour, and the control system. Then a direct inverse control scheme using the proposed SCAFIN is adopted to control the temperature of the RTP system. The SCAFIN is inherently a modified TSK-type fuzzy rule-based model possessing neural network's learning ability. There are no rules initially in the SCAFIN. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter identification. Simulation results show that the control approach is able to track a temporally varying temperature trajectory and maintain the uniformity of the spatial temperature distribution of the wafer in the RTP system simultaneously. &copy; 1999 Elsevier Science B.V. All rights reserved.
Juang, C.F. & Lin, C.T. 1999, 'A recurrent self-organizing neural fuzzy inference network', IEEE Transactions on Neural Networks, vol. 10, no. 4, pp. 828-845.
View/Download from: Publisher's site
A recurrent self-organizing neural fuzzy inference network (RSONFIN) is proposed in this paper. The RSONFIN is inherently a recurrent multilayered connectionist network for realizing the basic elements and functions of dynamic fuzzy inference, and may be considered to be constructed from a series of dynamic fuzzy rules. The temporal relations embedded in the network are built by adding some feedback connections representing the memory elements to a feedforward neural fuzzy network. Each weight as well as node in the RSONFIN has its own meaning and represents a special element in a fuzzy rule. There are no hidden nodes (i.e., no membership functions and fuzzy rules) initially in the RSONFIN. They are created on-line via concurrent structure identification (the construction of dynamic fuzzy if-then rules) and parameter identification (the tuning of the free parameters of membership functions). The structure learning together with the parameter learning forms a fast learning algorithm for building a small, yet powerful, dynamic neural fuzzy network. Two major characteristics of the RSONFIN can thus be seen: 1) the recurrent property of the RSONFIN makes it suitable for dealing with temporal problems and 2) no predetermination, like the number of hidden nodes, must be given, since the RSONFIN can find its optimal structure and parameters automatically and quickly. Moreover, to reduce the number of fuzzy rules generated, a flexible input partition method, the aligned clustering-based algorithm, is proposed. Various simulations on temporal problems are done and performance comparisons with some existing recurrent networks are also made. Efficiency of the RSONFIN is verified from these results. &copy; 1999 IEEE.
Lin, C.T., Juang, C.F. & Li, C.P. 1999, 'Temperature control with a neural fuzzy inference network', IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews, vol. 29, no. 3, pp. 440-451.
View/Download from: Publisher's site
Although multilayered backpropagation neural networks (BPNN's) have demonstrated high potential in the nonconventional branch of adaptive control, their long training time usually discourages their applications in industry. Moreover, when they are trained on-line to adapt to plant variations, the over-tuned phenomenon usually occurs. To overcome the weakness of the BPNN, in this paper we propose a neural fuzzy inference network (NFIN) suitable for adaptive control of practical plant systems in general and for adaptive temperature control of a water bath system in particular. The NFIN is inherently a modified Takagi-Sugeno-Kang (TSK)-type fuzzy rule-based model possessing a neural network's learning ability. In contrast to the general adaptive neural fuzzy networks, where the rules should be decided in advance before parameter learning is performed, there are no rules initially in the NFIN. The rules in the NFIN are created and adapted as on-line learning proceeds via simultaneous structure and parameter identification. The NFIN has been applied to a practical water bath temperature-control system. As compared to the BPNN under the same training procedure, the simulated results show that not only can the NFIN greatly reduce the training time and avoid the over-tuned phenomenon, but the NFIN also has perfect regulation ability. The performance of the NFIN is also compared to that of the traditional PID controller and fuzzy logic controller (FLC) on the water bath temperature-control system. The three control schemes are compared through experimental studies with respect to set-points regulation, ramp-points tracking, and the influence of unknown impulse noise and large parameter variation in the temperature-control system. It is found that the proposed NFIN control scheme has the best control performance of the three control schemes. &copy; 1999 IEEE.
Lin, C.T. & Kan, M.C. 1998, 'Adaptive fuzzy command acquisition with reinforcement learning', IEEE Transactions on Fuzzy Systems, vol. 6, no. 1, pp. 102-121.
View/Download from: Publisher's site
This paper proposes a four-layered adaptive fuzzy command acquisition network (AFCAN) for adaptively acquiring fuzzy command via interactions with the user or environment. It can catch the intended information from a sentence (command) given in natural language with fuzzy predicates. The intended information includes a meaningful semantic action and the fuzzy linguistic information of that action (for example, the phrase "move forward" represents the meaningful semantic action and the phrase "very high speed" represents the linguistic information in the fuzzy command "move forward at a very high speed"). The proposed AFCAN has three important features. First, we can make no restrictions whatever on the fuzzy command input, which is used to specify the desired information, and the network requires no acoustic, prosodic, grammar, and syntactic structure. Second, the linguistic information of an action is learned adaptively and it is represented by fuzzy numbers based on -level sets. Third, the network can learn during the course of performing the task. The AFCAN can perform off-line as well as on-line learning. For the off-line learning, the mutual-information (MI) supervised learning scheme and the fuzzy backpropagation (FBP) learning scheme are employed when the training data are available in advance. The former learning scheme is used to learn meaningful semantic actions and the latter learn linguistic information. The AFCAN can also perform on-line learning interactively when it is in use for fuzzy command acquisition. For the on-line learning, the MI-reinforcement learning scheme and the fuzzy reinforcement learning scheme are developed for the on-line learning of meaningful actions and linguistic information, respectively. An experimental system (fuzzy commands acquisition of a voice control system) is constructed to illustrate the performance and applicability of the proposed AFCAN. &copy; 1998 IEEE.
Juang, C.F. & Lin, C.T. 1998, 'An on-line self-constructing neural fuzzy inference network and its applications', IEEE Transactions on Fuzzy Systems, vol. 6, no. 1, pp. 12-32.
View/Download from: Publisher's site
A self-constructing neural fuzzy inference network (SONFIN) with on-line learning ability is proposed in this paper. The SONFIN is inherently a modified Takagi-Sugeno-Kang (TSK)-type fuzzy rule-based model possessing neural network's learning ability. There are no rules initially in the SONFIN. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter identification. In the structure identification of the precondition part, the input space is partitioned in a flexible way according to a aligned clustering-based algorithm. As to the structure identification of the consequent part, only a singleton value selected by a clustering method is assigned to each rule initially. Afterwards, some additional significant terms (input variables) selected via a projection-based correlation measure for each rule will be added to the consequent part (forming a linear equation of input variables) incrementally as learning proceeds. The combined precondition and consequent structure identification scheme can set up an economic and dynamically growing network, a main feature of the SONFIN. In the parameter identification, the consequent parameters are tuned optimally by either least mean squares (LMS) or recursive least squares (RLS) algorithms and the precondition parameters are tuned by backpropagation algorithm. Both the structure and parameter identification are done simultaneously to form a fast learning scheme, which is another feature of the SONFIN. Furthermore, to enhance the knowledge representation ability of the SONFIN, a linear transformation for each input variable can be incorporated into the network so that much fewer rules are needed or higher accuracy can be achieved. Proper linear transformations are also learned dynamically in the parameter identification phase of the SONFIN. To demonstrate the capability of the proposed SONFIN, simulations in different areas including control, communication, and signal processing are done....
Chen, P.H., Lai, J.H. & Lin, C.T. 1998, 'Application of fuzzy control to a road tunnel ventilation system', Fuzzy Sets and Systems, vol. 100, no. 1-3, pp. 9-28.
This paper deals with the serious problems of ventilation system in a large road tunnel. Higher visibility and lower concentration of carbon monoxide are the key issues concerning the ventilation system. Prior to designing the fuzzy control model, a configuration layout of the ventilation system including sensing, control and traffic prediction as well is conceptually constructed. Based on the layout that offers assignments of sensors and control elements, a fuzzy logic control model is developed. Membership functions of sensor errors and control increments are physically submitted in order to set up the fuzzy logic rules. Timing and spacing filtering in terms of weighting approaches is employed in the fuzzy logic rules. A dynamic equation describing the concentration of air pollution is also given so as to cooperate with the fuzzy logic rules and to play roles in the computer simulation. The result of computer simulation involving five cases indicates that a multi-level scheme is able to solve the engineering problems. &copy; 1998 Elsevier Science B.V. All rights reserved.
Lin, C.T., Jou, C.P. & Lin, C.J. 1998, 'GA-based reinforcement learning for neural networks', International Journal of Systems Science, vol. 29, no. 3, pp. 233-247.
A genetic reinforcement neural network (GRNN) is proposed to solve various reinforcement learning problems. The proposed GRNN is constructed by integrating two feedforward multilayer networks. One neural network acts as an action network for determining the outputs (actions) of the GRNN, and the other as a critic network to help the learning of the action network. Using the temporal difference prediction method, the critic network can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the action network. The action network uses the genetic algorithm (GA) to adapt itself according to the internal reinforcement signal. The key concept of the proposed GRNN learning scheme is to formulate the internal reinforcement signal as the fitness function for the GA. This learning scheme forms a novel hybrid GA, which consists of the temporal difference and gradient descent methods for the critic network learning, and the GA for the action network learning. By using the internal reinforcement signal as the fitness function, the GA can evaluate the candidate solutions (chromosomes) regularly, even during the period without external reinforcement feedback from the environment. Hence, the GA can proceed to new generations regularly without waiting for the arrival of the external reinforcement signal. This can usually accelerate the GA learning because a reinforcement signal may only be available at a time long after a sequence of actions has occurred in the reinforcement learning problems. Computer simulations have been conducted to illustrate the performance and applicability of the proposed learning scheme.
Wang, Y.J. & Lin, C.T. 1998, 'Runge-Kutta neural network for identification of dynamical systems in high accuracy', IEEE Transactions on Neural Networks, vol. 9, no. 2, pp. 294-307.
View/Download from: Publisher's site
This paper proposes the Runge-Kutta neural networks (RKNN's) for identification of unknown dynamical systems described by ordinary differential equations (i.e., ordinary differential equation or ODE systems) in high accuracy. These networks are constructed according to the Runge-Kutta approximation method. The main attraction of the RKNN's is that they precisely estimate the changing rates of system states (i.e., the right-hand side of the ODE x = f(x)) directly in their subnetworks based on the space-domain interpolation within one sampling interval such that they can do long-term prediction of system state trajectories. We show theoretically the superior generalization and long-term prediction capability of the RKNN's over the normal neural networks. Two types of learning algorithms are investigated for the RKNN's, gradient-and nonlinear recursive least-squares-based algorithms. Convergence analysis of the learning algorithms is done theoretically. Computer simulations demonstrate the proved properties of the RKNN's. &copy; 1998 IEEE.
Wang, Y.J. & Lin, C.T. 1998, 'A second-order learning algorithm for multilayer networks based on block Hessian matrix', Neural Networks, vol. 11, no. 9, pp. 1607-1622.
View/Download from: Publisher's site
This article proposes a new second-order learning algorithm for training the multilayer perceptron (MLP) networks. The proposed algorithm is a revised Newton's method. A forward-backward propagation scheme is first proposed for network computation of the Hessian matrix, H, of the output error function of the MLP. A block Hessian matrix, H(b), is then defined to approximate and simplify H. Several lemmas and theorems are proved to uncover the important properties of H and H(b), and verify the good approximation of H(b) to H; H(b) preserves the major properties of H. The theoretic analysis leads to the development of an efficient way for computing the inverse of H(b) recursively. In the proposed second-order learning algorithm, the least squares estimation technique is adopted to further lessen the local minimum problems. The proposed algorithm overcomes not only the drawbacks of the standard backpropagation algorithm (i.e. slow asymptotic convergence rate, bad controllability of convergence accuracy, local minimum problems, and high sensitivity to learning constant), but also the shortcomings of normal Newton's method used on the MLP, such as the lack of network implementation of H, ill representability of the diagonal terms of H, the heavy computation load of the inverse of H, and the requirement of a good initial estimate of the solution (weights). Several example problems are used to demonstrate the efficiency of the proposed learning algorithm. Extensive performance (convergence rate and accuracy) comparisons of the proposed algorithm with other learning schemes (including the standard backpropagation algorithm) are also made.
Lin, C.J. & Lin, C.T. 1997, 'An ART-based fuzzy adaptive learning control network', IEEE Transactions on Fuzzy Systems, vol. 5, no. 4, pp. 477-496.
View/Download from: Publisher's site
This paper addresses the structure and an associated on-line learning algorithm of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed fuzzy adaptive learning control network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. An on-line structure/parameter learning algorithm called FALCON-ART is proposed for constructing the FALCON dynamically. It combines the backpropagation learning scheme for parameter learning and the fuzzy ART algorithm for structure learning. The FALCON-ART has some important features. First of all, it partitions the input state space and output control space using irregular fuzzy hyperboxes according to the distribution of training data. In many existing fuzzy or neural fuzzy control systems, the input and output spaces are always partitioned into "grids." As the number of input/output variables increases, the number of partitioned grids will grow combinatorially. To avoid the problem of combinatorial growing of partitioned grids in some complex systems, the FALCON-ART partitions the input/output spaces in a flexible way based on the distribution of training data. Second, the FALCON-ART can create and train the FALCON in a highly autonomous way. In its initial form, there is no membership function, fuzzy partition, and fuzzy logic rule. They are created and begin to grow as the first training pattern arrives. Thus, the users need not give it any a priori knowledge or even any initial information on these. More notably, the FALCON-ART can on-line partition the input/output spaces, tune membership functions, find proper fuzzy logic rules, and annihilate redundant rules dynamically upon receiving on-line incoming training data. Computer simulations have been conducted to illustrate the performance and applicability of the proposed system. &copy; 1997 IEEE.
Lin, C.T. & Juang, C.F. 1997, 'An adaptive neural fuzzy filter and its applications', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 27, no. 4, pp. 635-656.
View/Download from: Publisher's site
A new kind of nonlinear adaptive filter, the adaptive neural fuzzy filter (ANFF), based upon a neural network's learning ability and fuzzy if-then rule structure, is proposed in this paper. The ANFF is inherently a feedforward multilayered connectionist network which can learn by itself according to numerical training data or expert knowledge represented by fuzzy if-then rules. The adaptation here includes the construction of fuzzy if-then rules (structure learning), and the tuning of the free parameters of membership functions (parameter learning). In the structure learning phase, fuzzy rules are found based on the matching of input-output clusters. In the parameter learning phase, a backpropagation-like adaptation algorithm is developed to minimize the output error. There are no hidden nodes (i.e., no membership functions and fuzzy rules) initially, and both the structure learning and parameter learning are performed concurrently as the adaptation proceeds. However, if some linguistic information about the design of the filter is available, such knowledge can be put into the ANFF to form an initial structure with hidden nodes. Two major advantages of the ANFF can thus be seen: 1) a priori knowledge can be incorporated into the ANFF which makes the fusion of numerical data and linguistic information in the filter possible; and 2) no predetermination, like the number of hidden nodes, must be given, since the ANFF can find its optimal structure and parameters automatically. Moreover, in contrast to traditional fuzzy systems where the input-output spaces are partitioned as grid type causing the combinatorial growing of fuzzy rules as the input-output dimensions increase, irregular partitions are done in the ANFF according to the distribution of training data so fewer fuzzy rules will be generated. To demonstrate the performance of the ANFF, two applications, the nonlinear channel equalization and the adaptive noise cancellation, are simulated. Efficiency and advant...
Juang, C.F. & Lin, C.T. 1996, 'Self-organizing neural fuzzy inference network for indentification and control', Journal of Control Systems and Technology, vol. 4, no. 4, pp. 269-280.
A Self-Organizing Neural Fuzzy Inference Network (SONFIN) with on-line learning ability is proposed in this paper. The SONFIN in inherently a fuzzy rule-based model possessing neural network's learning ability. In contrast to the general adaptive neural fuzzy networks, where the rules should be decided in advance before parameter learning is performed, there are no rules initially in the SONFIN. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter identification. In this structure identification of the precondition part, the input space is partitioned in a flexible way according to an aligned clustering-based algorithm. As to the structure identification of the consequent part, only a singleton value selected by a clustering method is assigned to each rule initially. Afterwards, some additional terms (a linear combination of input variables) are added to the consequent part when necessary. Furthermore, to enhance the knowledge representation ability of the SONFIN, a linear transformation for each input variable can be incorporated into the network so that much fewer rules are needed or higher accuracy can be achieved. Proper linear transformation are also learned dynamically in the parameter identification phase of the SONFIN. To demonstrate the capability of the proposed SONFIN, simulations in identification and control problems are done. Effectiveness of the SONFIN is verified from these simulations.
Lin, C.J. & Lin, C.T. 1996, 'Reinforcement learning for an ART-based fuzzy adaptive learning control network', IEEE Transactions on Neural Networks, vol. 7, no. 3, pp. 709-731.
View/Download from: Publisher's site
This paper proposes a reinforcement fuzzy adaptive learning control network (RFALCON) for solving various reinforcement learning problems. The proposed RFALCON is constructed by integrating two fuzzy adaptive learning control networks (FALCON's), each of which is a connectionist model with a feedforward multilayer network developed for the realization of a fuzzy controller. One FALCON performs as a critic network (fuzzy predictor), and the other as an action network (fuzzy controller). Using the temporal difference prediction method, the critic network can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the action network. The action network performs a stochastic exploratory algorithm to adapt itself according to the internal reinforcement signal. An ART-based reinforcement structure/parameter-learning algorithm is developed for constructing the RFALCON dynamically. During the learning process, both structure learning and parameter learning are performed simultaneously in the two FALCON's. The proposed RFALCON can construct a fuzzy control system dynamically and automatically through a reward/penalty signal (i.e., a "good" or "bad" signal). It is best applied to the learning environment, where obtaining exact training data is expensive. The proposed RFALCON has two important features. First, it reduces the combinatorial demands placed by the standard methods for adaptive linearization of a system. Second, the RFALCON is a highly autonomous system. Initially, there are no hidden nodes (i.e., no membership functions or fuzzy rules). They are created and begin to grow as learning proceeds. The RFALCON can also dynamically partition the input-output spaces, tune activation (membership) functions, and find proper network connection types (fuzzy rules). Computer simulations have been conducted to illustrate the performance and applicability of the proposed learning scheme. &copy; 1996 IEEE.
Lin, C.T. 1996, 'Adaptive subsethood for neural fuzzy control', International Journal of Systems Science, vol. 27, no. 10, pp. 937-955.
The paper extends Kosko's fuzzy measure of subsethood (Kosko 1992) to a measure of mutual subsethood or fuzzy equivalence. Gaussian or bell-shaped fuzzy sets then simplify the new measure and allow supervised learning to learn and tune the fuzzy rules. The gaussian sets act as nodes in neural fuzzy control networks and give a simple closed form for the measure of mutual subsethood. The new adaptive subsethood controller (ASC) system uses the network structure to store, learn, and tune fuzzy rules. Simulations show how the ASC system can control a model car, balance an inverted wedge, and control the ball and beam system.
Lin, C.T. & Lu, Y.C. 1996, 'A neural fuzzy system with fuzzy supervised learning', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 26, no. 5, pp. 744-763.
View/Download from: Publisher's site
A neural fuzzy system learning with fuzzy training data (fuzzy if-then rules) is proposed in this paper. This system is able to process and learn numerical information as well as linguistic information. At first, we propose a five-layered neural network for the connectionist realization of a fuzzy inference system. The connectionist structure can house fuzzy logic rules and membership functions for fuzzy inference. We use -level sets of fuzzy numbers to represent linguistic information. The inputs, outputs, and weights of the proposed network can be fuzzy numbers of any shape. Furthermore, they can be hybrid of fuzzy numbers and numerical numbers through the use of fuzzy singletons. Based on interval arithmetics, a fuzzy supervised learning algorithm is developed for the proposed system. It extends the normal supervised learning techniques to the learning problems where only linguistic teaching signals are available. The fuzzy supervised learning scheme can train the proposed system with desired fuzzy input-output pairs which are fuzzy numbers instead of the normal numerical values. With fuzzy supervised learning, the proposed system can be used for rule base concentration to reduce the number of rules in a fuzzy rule base. Simulation results are presented to illustrate the performance and applicability of the proposed system. &copy; 1996 IEEE.
Lin, C.T., Kan, M.C. & Chung, I.F. 1996, 'A neural network that learns from fuzzy data for language acquisition', International Journal of Uncertainty, Fuzziness and Knowlege-Based Systems, vol. 4, no. 6, pp. 581-603.
This paper proposes a four-layered fuzzy language acquisition network (FLAN) for acquiring fuzzy language. It can catch the intended information from a sentence (command) spoken in natural language with fuzzy terms. The intended information includes a meaningful semantic action and the fuzzy linguistic information of that action (for example, the phrase "move forward" represents the meaningful semantic action and the phrase "very high speed" represents the linguistic information in the fuzzy command "Move forward in a very high speed."). The proposed FLAN has two important features. First, we can make no restrictions whatever on the fuzzy language input which is used to specify the desired information, and the network requires no acoustic, prosodic, grammar and syntactic structure. Second, the linguistic information of an action is learned automatically and it is represented by fuzzy numbers based on -level sets. A supervised learning scheme is proposed to train the FLAN on fuzzy training data. This learning scheme consists of the mutual-information (MI) supervised learning algorithm for learning meaningful semantic actions, and the fuzzy backpropagation (FBP) learning algorithm for learning linguistic information. An experimental system is constructed to illustrate the performance and applicability of the proposed FLAN.
Lin, C.T. & Lee, C.S.G. 1995, 'A Multi-Valued Boltzmann Machine', IEEE Transactions on Systems, Man, and Cybernetics, vol. 25, no. 4, pp. 660-669.
View/Download from: Publisher's site
The idea of Hopfield network is based on the Ising spin glass model in which each spin has only two possible states: up and down. By introducing stochastic factors into this network and performing a simulated annealing process on it, it becomes a Boltzmann machine which can escape from local minimum states to achieve the global minimum. This paper generalizes the above ideas to multi-value case based on the XY spin glass model in which each spin can be in any direction in a plane. Simply using the gradient descent method and the analog Hopfield network, two different analog connectionist structures and their corresponding evolving rules are first designed to transform the XY spin glass model to distributed computational models. These two analog computational models are single-layered connectionist structures and multi-layered Hopfield analog networks. The latter network eases the node (neuron) computational requirement of the former at the expense of more neurons and connections. With the proposed evolving rules, the proposed models evolve according to a predefined Hamiltonian (energy function) which will decrease until it reaches a (perhaps local) minimum. Since these two structures can easily get stuck in local minima, a multivalued Boltzmann machine is proposed which adopts the discrete planar spin glass model for the local minimum problem. Each neuron in the multi-valued Boltzmann machine can only take n discrete directions (states). The stochastic simulated annealing method is introduced to the evolving rules of the multi-valued Boltzmann machine to solve the local minimum problem. The multi-valued Boltzmann machine can be applied to the mobile robot navigation problem by defining proper artificial magnetic field on the traverse terrain. This new artificial magnetic field approach for the mobile robot navigation problem has shown to have several advantages over existing graph search and potential field techniques. &copy; 1995 IEEE
Lin, C.T. & Lu, Y.C. 1995, 'A Neural Fuzzy System with Linguistic Teaching Signals', IEEE Transactions on Fuzzy Systems, vol. 3, no. 2, pp. 169-189.
View/Download from: Publisher's site
A neural fuzzy system learning with linguistic teaching signals is proposed in this paper. This system is able to process and learn numerical information as well as linguistic information. It can be used either as an adaptive fuzzy expert system or as an adaptive fuzzy controller. At first, we propose a fivelayered neural network for the connectionist realization of a fuzzy inference system. The connectionist structure can house fuzzy logic rules and membership functions for fuzzy inference. We use a-level sets of fuzzy numbers to represent linguistic information. The inputs, outputs, and weights of the proposed network can be fuzzy numbers of any shape. Furthermore, they can be hybrid of fuzzy numbers and numerical numbers through the use of fuzzy singletons. Based on interval arithmetics, two kinds of learning schemes are developed for the proposed system: fuzzy supervised learning and fuzzy reinforcement learning. They extend the normal supervised and reinforcement learning techniques to the learning problems where only linguistic teaching signals are available. The fuzzy supervised learning scheme can train the proposed system with desired fuzzy input-output pairs which are fuzzy numbers instead of the normal numerical values. With fuzzy supervised learning, the proposed system can be used for rule base concentration to reduce the number of rules in a fuzzy rule base. In the fuzzy reinforcement learning problem that we consider, the reinforcement signal from the environment is linguistic information (fuzzy critic signal) such as 'good, 'very good, or 'bad, instead of the normal numerical critic values such as '0" (success) or 1" (failure). With the fuzzy critic signal from the environment, the proposed system can learn proper fuzzy control rules and membership functions. We discuss two kinds of fuzzy reinforcement learning problems: singlestep prediction problems and multistep prediction problems. Simulation results are presented to illustrate the performa...
Lin, C.T. 1995, 'A neural fuzzy control system with structure and parameter learning', Fuzzy Sets and Systems, vol. 70, no. 2-3, pp. 183-212.
View/Download from: Publisher's site
A general connectionist model, called neural fuzzy control network (NFCN), is proposed for the realization of a fuzzy logic control system. The proposed NFCN is a feedforward multilayered network which integrates the basic elements and functions of a traditional fuzzy logic controller into a connectionist structure which has distributed learning abilities. The NFCN can be constructed from supervised training examples by machine learning techniques, and the connectionist structure can be trained to develop fuzzy logic rules and find membership functions. Associated with the NFCN is a two-phase hybrid learning algorithm which utilizes unsupervised learning schemes for structure learning and the backpropagation learning scheme for parameter learning. By combining both unsupervised and supervised learning schemes, the learning speed converges much faster than the original backpropagation algorithm. The two-phase hybrid learning algorithm requires exact supervised training data for learning. In some real-time applications, exact training data may be expensive or even impossible to obtain. To solve this problem, a reinforcement neural fuzzy control network (RNFCN) is further proposed. The RNFCN is constructed by integrating two NFCNs, one functioning as a fuzzy predictor and the other as a fuzzy controller. By combining a proposed on-line supervised structure-parameter learning technique, the temporal difference prediction method, and the stochastic exploratory algorithm, a reinforcement learning algorithm is proposed, which can construct a RNFCN automatically and dynamically through a reward-penalty signal (i.e., "good" or "bad" signal). Two examples are presented to illustrate the performance and applicability of the proposed models and learning algorithms. &copy; 1995.
Lin, C.J. & Lin, C.T. 1995, 'Adaptive fuzzy control of unstable nonlinear systems.', International journal of neural systems, vol. 6, no. 3, pp. 283-298.
This paper addresses the structure and an associated on-line learning algorithm of a feedforward multilayer connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed Fuzzy Adaptive Learning Control Network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. An on-line structure/parameter learning algorithm, called FALCON-ART, is proposed for constructing the FALCON dynamically. The FALCON-ART can partition the input/output space in a flexible way based on the distribution of the training data. Hence it can avoid the problem of combinatorial growing of partitioned grids in some complex systems. It combines the backpropagation learning scheme for parameter learning and the fuzzy ART algorithm for structure learning. More notably, the FALCON-ART can on-line partition the input/output spaces, tune membership functions, and find proper fuzzy logic rules dynamically without any a priori knowledge or even any initial information on these. The proposed learning scheme has been successfully used to control two unstable nonlinear systems. They are the seesaw system and the inverted wedge system.
Lin, C.T., Lin, C.J. & Lee, C.S.G. 1995, 'Fuzzy adaptive learning control network with on-line neural learning', Fuzzy Sets and Systems, vol. 71, no. 1, pp. 25-45.
View/Download from: Publisher's site
This paper addresses the structure and the associated on-line learning algorithms of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed Fuzzy Adaptive Learning COntrol Network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. The connectionist structure of the proposed FALCON can be constructed from training examples by neural learning techniques to find proper fuzzy partitions, membership functions, and fuzzy logic rules. Two complementary on-line structure/parameter learning algorithms, FALCON-FSM and FALCON-ART, are proposed for constructing the FALCON dynamically. The FALCON-FSM combines the backpropagation learning scheme for parameter learning and a fuzzy similarity measure for structure learning. The FALCON-FSN can find proper fuzzy logic rules, membership functions, and the size of output partitions simultaneously. In the FALCON-FSM algorithm, the input and output spaces are partitioned into "grids". The grid-typed space partitioning certainly makes both the fuzzy logic controller software emulation and fuzzy chip implementation convenient. However, as the number of input/output variables increases, the number of partitioned grids will grow combinatorially. To avoid the problem of combinatorial growth of partitioned grids in some complex systems, the FALCON-ART algorithm is developed, which can partition the input and output spaces in a more flexible way based on the distribution of the training data. The FALCON-ART combines the backpropagation learning scheme for parameter learning and a fuzzy ART algorithm for structure learning. The FALCON-ART can on-line partition the input and output spaces, tune membership functions and find proper fuzzy logic rules dynamically. Computer simulations were conducted to illustrate the performance and applicability of both FALCON-FSM and FALCON-ART ...
Hsiao, C.H., Lin, C.T. & Cassidy, M. 1994, 'Application of fuzzy logic and neural networks to automatically detect freeway traffic incidents', Journal of Transportation Engineering, vol. 120, no. 5, pp. 753-772.
View/Download from: Publisher's site
To date, efforts to manage freeway congestion have been seriously impeded by the inability to promptly and reliably detect the presence of traffic incidents. Traditional incident-detection algorithms distinguish between congested and uncongested operation by comparing measured traffic-stream parameters with predefined threshold values. Given the range of possible operating conditions in the traffic stream, selecting a single threshold value, and the suitability of that selected threshold, is full of uncertainty. This inherent uncertainty makes fuzzy logic a promising approach to incident detection. A Fuzzy Logic Incident Patrol System (FLIPS) is proposed to solve many of the problems inherent in traditional incident-detection algorithms. The FLIPS combines fuzzy logic with the learning capabilities of neural networks to form a connectionist model. The system can be constructed automatically from training examples to find the optimal input/output membership functions. Threshold values, implicitly obtained by fuzzy-logic rules and membership functions, are treated as dependent variables, which change according to prevailing traffic-stream parameters measured by detectors. The FLIPS avoids the rule-matching time of the inference engine in the traditional fuzzy-logic system. The potential effectiveness of the FLIPS is evaluated using an empirical database collected in Toronto, Canada. Future refinement to the FLIPS are also discussed in this paper. &copy; ASCE.
Lin, C.T. & George Lee, C.S. 1994, 'Reinforcement Structure/Parameter Learning for Neural-Network-Based Fuzzy Logic Control Systems', IEEE Transactions on Fuzzy Systems, vol. 2, no. 1, pp. 46-63.
View/Download from: Publisher's site
This paper proposes a reinforcement neural-network-based fuzzy logic control system (RNN-FLCS) for solving various reinforcement learning problems. The proposed RNN-FLCS is constructed by integrating two neural-network-based fuzzy logic controllers (NN-FLC's), each of which is a connectionist model with a feedforward multilayered network developed for the realization of a fuzzy logic controller. One NN-FLC performs as a fuzzy predictor, and the other as a fuzzy controller. Using the temporal difference prediction method, the fuzzy predictor can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the fuzzy controller. The fuzzy controller performs a stochastic exploratory algorithm to adapt itself according to the internal reinforcement signal. During the learning process, both structure learning and parameter learning are performed simultaneously in the two NN-FLC's using the fuzzy similarity measure. The proposed RNN-FLCS can construct a fuzzy logic control and decision-making system automatically and dynamically through a reward/penalty signal (i.e., a 'good or 'bad signal) or through very simple fuzzy information feedback such as 'high, 'too high, 'low, and 'too low. The proposed RNN-FLCS is best applied to the learning environment, where obtaining exact training data is expensive. The proposed RNN-FLCS also preserves the advantages of the original NN-FLC, such as the ability to find proper network structure and parameters simultaneously and dynamically and to avoid the rule-matching time of the inference engine in the traditional fuzzy logic systems. Computer simulations were conducted to illustrate the performance and applicability of the proposed RNN-FLCS. &copy; 1994 IEEE
Lin, C.T. & Lee, G.C.S. 1991, 'Neural-Network-Based Fuzzy Logic Control and Decision System', IEEE Transactions on Computers, vol. 40, no. 12, pp. 1320-1336.
View/Download from: Publisher's site
A general neural-network (connectionist) model for fuzzy logic control and decision/diagnosis systems is proposed. The proposed connectionist model can be contrasted with the traditional fuzzy logic control and decision system in their network structure and learning ability. Such fuzzy control/decision networks can be constructed from training examples by machine learning techniques, and the connectionist structure can be trained to develop fuzzy logic rules and find optimal input/output membership functions. By combining both unsupervised (self-organized) and supervised learning schemes, the learning speed converges much faster than the original backpropagation learning algorithm. This connectionist model also provides human-understandable meaning to the normal feedforward multilayer neural network in which the internal units are always opaque to the users. The connectionist structure also avoids the rule-matching time of the inference engine in the traditional fuzzy logic system. Two examples are presented to illustrate the performance and applicability of the proposed connectionist model. &copy; 1991 IEEE
Lin, C.T. & Lee, C.S.G. 1991, 'Fault-Tolerant Reconfigurable Architecture for Robot Kinematics and Dynamics Computations', IEEE Transactions on Systems, Man and Cybernetics, vol. 21, no. 5, pp. 983-999.
View/Download from: Publisher's site
The computations of kinematics, dynamics, Jacobian, and their corresponding inverses are six essential problems in the control of robot manipulators. Efficient parallel algorithms for these computations are discussed and analyzed, and their characteristics are identified based on type of parallelism, degree of parallelism, uniformity of the operations, fundamental operations, data dependencies, and communication requirement. It is shown that most of the algorithms for robotic computations possess highly regular properties and some common structures, especially the linear recursive structure. They are well-suited to be implemented on a single-instruction-stream multiple-data-stream (SIMD) computer with reconfigurable interconnection networks. A reconfigurable, dual-network, SIMD (DN-SIMD) machine with internal direct feedback that best matches these characteristics has been designed. To achieve high efficiency in the computations of robotic algorithms on the proposed parallel machine, a generalized cube interconnection network is proposed and investigated. A centralized network switch control scheme is also developed to support the pipeline timing of this machine. Moreover, to maintain a high reliability in the overall system, a fault-tolerant generalized cube (FTGC) network is designed to improve the original network. With this improvement, it is shown that the proposed parallel architecture performs correctly even under a single fault condition including a processing element fault or a network component fault. The multiple fault tolerance of FTGC network is also discussed. The use of the DN-SIMD machine is illustrated through an example of the inverse dynamics computation. &copy; 1991 IEEE