Global-Leading Research and Education in Brain and Cognitive engineering KOREA UNIVERSITY To understand the structure and the key principles of high-order information processing by which brains work,
and to implement these in artificial systems that interact intelligently with real world

Research Highlight

Research
Prof. Seong-Whan Lee’s paper published in IEEE Transactions on Pattern Analysis and Machine Intelligence
Author Administrator (IP: *.152.163.70) Date 2012-05-18 02:32 Views 528
원본사진보기

Prof. Seong-Whan Lee’s new research on `A Novel Bayesian Framework for Discriminative Feature Extraction in Brain-Computer Interfaces` was accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence (IF top 1%)

This paper receives credit for contribution by proposing a framework which can train and optimize spatio-spectral filter simultaneously for classifying EEG signals generated by motor imageries. In the framework, the problem of simultaneous spatio-spectral filter optimization is formulated as the estimation of an unknown posterior pdf by representing frequency bands as continuous random variables. Also, the Modified Factored-Sampling method and an observation model to quantitatively measure levels of class discrimination of feature vectors are proposed in the paper. The proposed methods are the first probabilistic approach to solve the problem of simultaneous spatio-spectral filter optimization and have significant meanings with automatic data-driven learning for various frequency bands and corresponding spatial filters. From the viewpoint of classifier design, the proposed method naturally allows us to construct a spectrally weighted label decision rule by linearly combining the outputs from multiple classifiers. The authors demonstrated the feasibility and effectiveness of the proposed methods by analyzing the results on international public databases.

Please refer to the following title and abstract.

Title: A Novel Bayesian Framework for Discriminative Feature Extraction in Brain-Computer Interfaces

Abstract:
As there has been a paradigm shift in the learning load from a human subject to a computer, machine learning has been considered as a useful tool for Brain-Computer Interfaces (BCIs). In this paper, we propose a novel Bayesian framework for discriminative feature extraction for motor imagery classification in an EEG-based BCI, in which the class-discriminative frequency bands and the corresponding spatial filters are optimized by means of the probabilistic and information-theoretic approaches. In our framework, the problem of simultaneous spatio-spectral filter optimization is formulated as the estimation of an unknown posterior pdf that represents the probability that a single-trial EEG of predefined mental tasks can be discriminated in a state. In order to estimate the posterior pdf, we propose a particle-based approximation method by extending a factored-sampling technique with a diffusion process. An information-theoretic observation model is also devised to measure discriminative power of features between classes. From the viewpoint of classifier design, the proposed method naturally allows us to construct a spectrally weighted label decision rule by linearly combining the outputs from multiple classifiers. We demonstrate the feasibility and effectiveness of the proposed method by analyzing the results and its success on three public databases
.

 

File re_img.gif(7.3K)


Research
Prof. Seong-Whan Lee`s paper published in Pattern Recognition
Prof. Seong-Whan Lee`s paper published in Pattern Recognition
Prof. Seong-Whan Lee’s new research on `Adaptive Occlusion State Estimation for Human Pose Tracking under Self-Occlusions` was accepted in Pattern Recognition (IF top 10%) In this paper, they proposed a method to solve self-occlusion problem in 3D human pose tracking. They modeled an “occlusion relationship (= state)” between human body parts as a graph and proposed a novel “adaptive occlusion state estimation” method, which infer human body con…

2012-10-25

Prof. Seong-Whan Lee`s paper published in Pattern Recognition
Prof. Seong-Whan Lee`s paper published in Pattern Recognition
Prof. Seong-Whan Lee’s new research on `Person Authentication from Neural Activity of Face-Specific Visual Self-Representation` was accepted in Pattern Recognition (IF top 10%) This paper has been contributed to EEG based new biometric system construction based on the neurophysiological features of face-specific visual self representation in a human brain. In particular, unlike previous methods that considered the brain activities of the resting state,…

2012-10-25

Prof. Jae-Ho Han’s paper published in IEEE Journal of Selected Topics in Quantum Electronics
Prof. Jae-Ho Han’s paper published in IEEE Journal of Selected Topics in Quant…
Prof. Jaeho Han’s new research on `Dichloromethane Detection Based on Near Infrared Absorptive Sensing` was published in IEEE Journal of Selected Topics in Quantum Electronics. This paper demonstrated a biophotonic near-infrared sensor using a tunable fiber ring laser based on an absorption spectroscopic method. A continuous wave near-infrared source operating in the C-band wavelength range has been developed and the Sagnac loop mirror, which operates a…

2012-08-10

Prof. Jichai Jeong`s paper published in Journal of Lightwave Technology
Prof. Jichai Jeong`s paper published in Journal of Lightwave Technology
Prof. Jichai Jeong’s new research on ‘DFT-Precoded Coherent Optical OFDM with Hermitian Symmetry for Fiber Nonlinearity Mitigation’ was published in Journal of Lightwave Technology

2012-08-10

Prof. Anil Jain`s paper published in IEEE Transactions on Pattern Analysis and Machine Intelligence
Prof. Anil Jain`s paper published in IEEE Transactions on Pattern Analysis and M…
Prof. Anil Jain’s new research on `Altered Fingerprints: Analysis and Detection` was published in IEEE Transactions on Pattern Analysis and Machine Intelligence (IF top 1%)

2012-08-10

Prof. Jeehyun Kwag’s paper published in Neuropharmacology
Prof. Jeehyun Kwag’s paper published in Neuropharmacology
Prof. Jeehyun Kwag’s new research on `Gating of NMDA Receptor-Mediated Hippocampal Spike Timing-Dependent Potentiation by MGluR5` was published in Neuropharmacology This paper demonstrates that the timing long-term poten hippocampal synapses. The induction of tLTP was blocked antagonist and by the selective mGluR5 blockers, MPEP(2 ((2-Methyl-4-thiazolyl)ethynyl)pyridine), whereas the activation of Dihydroxyphenylglycine(DHPG) facilitated the induction …

2012-08-10

Prof. Seong-Whan Lee awarded the
Prof. Seong-Whan Lee awarded the "Scientist of the Month Award"
Professor Seong-Whan Lee of the Korea University received the Scientist of the Month Award from the Ministry of Education, Science and Technology (Minister Jooho Lee) and the National Research Foundation of Korea (President Seungjong Lee) for developing the ingenious technology that a computer can automatically recognize the human sign language and facial expressions, etc. Professor Lee is a renowned scholar in the field of pattern recognition techniques…

2012-05-18

Prof. Seong-Whan Lee’s paper published in IEEE Transactions on Pattern Analysis and Machine Intelligence
Prof. Seong-Whan Lee’s paper published in IEEE Transactions on Pattern Analysi…
Prof. Seong-Whan Lee’s new research on `A Novel Bayesian Framework for Discriminative Feature Extraction in Brain-Computer Interfaces` was accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence (IF top 1%) This paper receives credit for contribution by proposing a framework which can train and optimize spatio-spectral filter simultaneously for classifying EEG signals generated by motor imageries. In the framework, the problem of sim…

2012-05-18

Prof. Alan Yuille’s paper published in IEEE Transactions on Pattern Analysis and Machine Intelligence
Prof. Alan Yuille’s paper published in IEEE Transactions on Pattern Analysis a…
Prof. Alan Yuille’s new research on `Recursive Segmentation and Recognition Templates for Image Parsing` was accepted in IEEE Transactions on Pattern Analysis and Machine Intelligence (IF top 1%) In this paper, they proposed a Hierarchical Image Model (HIM) which parses images to perform segmentation and object recognition. The HIM represents the image recursively by segmentation and recognition templates at multiple levels of the hierarchy. The HIM ha…

2012-05-18

Prof. Jaeseok Park`s paper published in NeuroImage
Prof. Jaeseok Park`s paper published in NeuroImage
Prof. Jaeseok Park`s new research on `Susceptibility-Resistant Variable-Flip-Angle Turbo Spin Echo Imaging for Reliable Estimation of Cortical Thickness: A Feasibility Study` was published in NeuroImage In this paper, a novel magnetic resonance imaging pulse sequence was developed as an alternative to  conventional imaging methods for brain morphological studies that yield inaccurate estimation of cortical thickness resulting from susceptibility-ind…

2012-05-18

 1 2 3 4 5 6 7 8 9