Sang Wan Lee

Dynamic-footprint based person identification using mat-type pressure sensor

Many diverse methods have been developing in the field of biometric identification as human-friendliness has been emphasized in the intelligent systems area. And one of emerging method is to use human walking behavior. But, in the previous methods based on human gait, stable somewhat long-term walking data are an essential condition for person recognition. Therefore, these methods are difficult to cope with various change of walking velocity which may be generated frequently during real walking. In this paper, we suggest a new method which uses just one-step walking data from mat-type pressure sensor. When a human walk through the pressure sensor, we get quantized COP (center of pressure) trajectory and HMM (hidden Markov model) is used to make probability models for users each foot. And then, HMMs for two feet are combined for better performance by Levenberg-Marquart learning method. Finally, we prove the usefulness of the suggested method using 8 people recognition experiments.

Facial Emotional Expression Recognition with Soft Computing Techniques

The facial expression recognition (FER) is one of the biosignal-based recognition techniques which attract a lot of attention recently. To deal with its complex characteristics effectively, we adopt the soft computing techniques (SCT) such as fuzzy logic, neural networks, genetic algorithm and/or rough set technique. In this paper, we overview the state-of-the-art reports on FER in view of SCT, and introduce some interesting works done by our group on the SCT-based facial emotional expression recognition. Specifically, 1) fuzzy observer-based approach, 2) personalized FER system based on fuzzy neural networks, and 3) Gabor wavelet neural networks are briefly discussed.

A Nonsupervised Learning Framework of Human Behavior Patterns Based on Sequential Actions

In designing autonomous service systems such as assistive robots for the aged and the disabled, discovery and prediction of human actions are important and often crucial. Patterns of human behavior, however, involve ambiguity, uncertainty, complexity, and inconsistency caused by physical, logical, and emotional factors, and thus their modeling and recognition are known to be difficult. In this paper, a nonsupervised learning framework of human behavior patterns is suggested in consideration of human behavioral characteristics. Our approach consists of two steps. In the first step, a meaningful structure of data is discovered by using Agglomerative Iterative Bayesian Fuzzy Clustering (AIBFC) with a newly proposed cluster validity index. In the second step, the sequence of actions is learned on the basis of the structure discovered in the first step and by utilizing the proposed Fuzzy-state Q--learning (FSQL) process. These two learning steps are incorporated in an amalgamated framework, AIBFC-FSQL, which is capable of learning human behavior patterns in a nonsupervised manner and predicting subsequent human actions. Through a number of simulations with typical benchmark data sets, we show that the proposed learning method outperforms several well-known methods. We further conduct experiments with two challenging real-world databases to demonstrate its usefulness from a practical perspective.

Robust EMG pattern recognition to muscular fatigue effect for powered wheelchair control

The main goal of this paper is to design an electromyogram (EMG) pattern classifier which is robust against muscular fatigue effects for powered wheelchair control. When a user operates a powered wheelchair using EMG-based interface for a long time, muscular fatigue often arises from sustained duration of muscle contraction. The recognition rate thus is degraded and controlling wheelchair gets more difficult. In this paper, an important observation is addressed that the variations of feature values due to the effect of the muscular fatigue are consistent for sustained duration. Based on this observation, we design a robust pattern classifier through the adaptation process of hyperboxes of Fuzzy Min-Max Neural Network. We present, as a result, a significantly improved performance in terms of the continuous usage of wheelchair.

Neural computations mediating one-shot learning in the human brain.

Incremental learning, in which new knowledge is acquired gradually through trial and error, can be distinguished from one-shot learning, in which the brain learns rapidly from only a single pairing of a stimulus and a consequence. Very little is known about how the brain transitions between these two fundamentally different forms of learning. Here we test a computational hypothesis that uncertainty about the causal relationship between a stimulus and an outcome induces rapid changes in the rate of learning, which in turn mediates the transition between incremental and one-shot learning. By using a novel behavioral task in combination with functional magnetic resonance imaging (fMRI) data from human volunteers, we found evidence implicating the ventrolateral prefrontal cortex and hippocampus in this process. The hippocampus was selectively “switched” on when one-shot learning was predicted to occur, while the ventrolateral prefrontal cortex was found to encode uncertainty about the causal association, exhibiting increased coupling with the hippocampus for high-learning rates, suggesting this region may act as a “switch,” turning on and off one-shot learning as required.

Feature subset selection using separability index matrix

Effective Feature Subset Selection (FSS) is an important step when designing engineering systems that classify complex data in real time. The electromyographic (EMG) signal-based walking assistance system is a typical system that requires an efficient computational architecture for classification. The performance of such a system depends largely on a criterion function that assesses the quality of selected feature subsets. However, many well-known conventional criterion functions use less relevant features for classification or they have a high computational cost. Here, we propose a new criterion function that provides more effective FSS. The proposed criterion function, known as a separability index matrix (SIM), provides features pertinent to the classification task and a very low computational cost. This new function produces to a simple feature selection algorithm when combined with the forward search paradigm. We performed extensive experimental comparisons in terms of classification accuracy and computational costs to confirm that the proposed algorithm outperformed other filter-type feature selection methods that are based on various distance measures, including inter-intra, Euclidean, Mahalanobis, and Bhattacharyya distances. We then applied the proposed method to a gait phase recognition problem in our EMG signal-based walking assistance system. We demonstrated that the proposed method performed competitively when compared with other wrapper-type feature selection methods in terms of class-separability and recognition rate.

Applying human learning principles to user-centered IoT systems

IoT systems can benefit from a process model based on principles derived from the psychology and neuroscience of human behavior that emulates how humans acquire task knowledge and learn to adapt to changing context.

Hand Gesture Recognition Using Multivariate Fuzzy Decision Tree and User Adaptation

As an emerging human-computer interaction HCI technology, recognition of human hand gesture is considered a very powerful means for human intention reading. To construct a system with a reliable and robust hand gesture recognition algorithm, it is necessary to resolve several major difficulties of hand gesture recognition, such as inter-person variation, intra-person variation, and false positive error caused by meaningless hand gestures. This paper proposes a learning algorithm and also a classification technique, based on multivariate fuzzy decision tree MFDT. Efficient control of a fuzzified decision boundary in the MFDT leads to reduction of intra-person variation, while proper selection of a user dependent UD recognition model contributes to minimization of inter-person variation. The proposed method is tested first by using two benchmark data sets in UCI Machine Learning Repository and then by a hand gesture data set obtained from 10 people for 15 days. The experimental results show a discernibly enhanced classification performance as well as user adaptation capability of the proposed algorithm.

Representation of a Fisher Criterion Function in a Kernel Feature Space

In this brief, we consider kernel methods for classification (Shawe-Taylor and Cristianini, 2004) from a separability point of view and provide a representation of the Fisher criterion function in a kernel feature space. We then show that the value of the Fisher function can be simply computed by using averages of diagonal and off-diagonal blocks of a kernel matrix. This result further serves to reveal that the ideal kernel matrix is a global solution to the problem of maximizing the Fisher criterion function. Its relation to an empirical kernel target alignment is then reported. To demonstrate the usefulness of these theories, we provide an application study for classification of prostate cancer based on microarray data sets. The results show that the parameter of a kernel function can be readily optimized.

Design of a Gait Phase Recognition System That Can Cope With EMG Electrode Location Variation

Electromyogram (EMG) signal-based gait phase recognition for walking-assist devices warrants much attention in human-centered system design as it well exemplifies human-in-the-loop control where the systems prediction directly affects subsequent walking motion. Since walking motion poses considerable variations in electrode placement, performance reliability of such systems is contingent on a combination of electrode montage and a feature extraction method that takes into account underlying physiological factors of peripheral muscles where electrodes are placed. In many practical applications, however, proper consideration of effects of the electrode location variation on performance reliability of the system has received scant empirical attention. Here, based on a user-centered design principle, we establish a gait phase recognition system that is capable of rigidly controlling ill effects due to this covariate by carrying out a large-scale analysis that combines statistical, model-based, and empirical approaches. In doing so, we have developed a special sensing suit for the control of electrode placement and a reliable data acquisition. We then have conducted a nonparametric statistical analysis on class separability values of thirty types of EMG feature sets, followed by a model-based analysis to address the tradeoff between class separability and dimensionality. To further address the issue of how these results generalize to independent systems and data sets, we have carried out an empirical performance assessment over six classification methods. First, the two feature types, Integral of Absolute Value and Histogram, and a combination of the two are shown to be robust against electrode location variations while providing a firm performance guarantee. Second, system organization scenarios are presented on a case-by-case basis, allowing us to trade off system complexity for on-line adaptation capability. Collectively, our integrated analysis lends itself to formulating a guideline for design of highly reliable EMG signal-based walking assistant systems in a variety of smart home scenarios.