Novel Multi-layer Non-negative Tensor Factorization with Sparsity Constraints

Cichocki, A.,Zdunek, R.,Choi, S.,Plemmons, R.,Amari, S.-i. SPRINGER-VERLAG 2007 ICHIT 2012 Vol.4432 No.-

NONNEGATIVE TENSOR FACTORIZATION FOR CONTINUOUS EEG CLASSIFICATION

이혜경(Hyekyoung Lee),A. Cichocki,최승진(Seungjin Choi) 한국정보과학회 2007 한국정보과학회 학술발표논문집 Vol.34 No.2A

GOM-Face: GKP, EOG, and EMG-Based Multimodal Interface With Application to Humanoid Robot Control

Nam, Yunjun,Koo, Bonkon,Cichocki, Andrzej,Choi, Seungjin IEEE 2014 IEEE Transactions on Biomedical Engineering Vol.61 No.2

<P>We present a novel human-machine interface, called GOM-Face , and its application to humanoid robot control. The GOM-Face bases its interfacing on three electric potentials measured on the face: 1) glossokinetic potential (GKP), which involves the tongue movement; 2) electrooculogram (EOG), which involves the eye movement; 3) electromyogram, which involves the teeth clenching. Each potential has been individually used for assistive interfacing to provide persons with limb motor disabilities or even complete quadriplegia an alternative communication channel. However, to the best of our knowledge, GOM-Face is the first interface that exploits all these potentials together. We resolved the interference between GKP and EOG by extracting discriminative features from two covariance matrices: a tongue-movement-only data matrix and eye-movement-only data matrix. With the feature extraction method, GOM-Face can detect four kinds of horizontal tongue or eye movements with an accuracy of 86.7% within 2.77 s. We demonstrated the applicability of the GOM-Face to humanoid robot control: users were able to communicate with the robot by selecting from a predefined menu using the eye and tongue movements.</P>

Tongue-Rudder: A Glossokinetic-Potential-Based Tongue–Machine Interface

Nam, Yunjun,Zhao, Qibin,Cichocki, Andrzej,Choi, Seungjin IEEE 2012 IEEE Transactions on Biomedical Engineering Vol.59 No.1

<P>Glossokinetic potentials (GKPs) are electric potential responses generated by tongue movement. In this study, we use these GKPs to automatically detect and estimate tongue positions, and develop a tongue-machine interface. We show that a specific configuration of electrode placement yields discriminative GKPs that vary depending on the direction of the tongue. We develop a linear model to determine the direction of tongue from GKPs, where we seek linear features that are robust to a baseline drift problem by maximizing the ratio of intertask covariance to intersession covariance. We apply our method to the task of wheelchair control, developing a tongue-machine interface for wheelchair control, referred to as tongue-rudder. A teeth clenching detection system, using electromyography, was also implemented in the system in order to assign teeth clenching as the stop command. Experiments on off-line cursor control and online wheelchair control confirm the unique advantages of our method, such as: 1) noninvasiveness, 2) fine controllability, and 3) ability to integrate with other EEG-based interface systems.</P>

NMF for Motor Imagery EEG Classification

이혜경(Hyekyoung Lee),Andrezej Cichocki,최승진(Seungjin Choi) 한국정보과학회 2006 한국정보과학회 학술발표논문집 Vol.33 No.1

In this paper, we present a method of feature extraction for motor imagery single trial EEG classification, where we exploit nonnegative matrix factorization (NMF) to select discriminative features in the time-frequency representation of EEG. Experimental results with motor imagery EEG data in BCI competition 2003, show that the method indeed finds meaningful EEG features automatically, while some existing methods should undergo cross-validation to find them.

CO₂Emission Reduction-the Contribution of modern HSDI Engines

Johannes Bierbaumer,Rainer Cichocki,Walter Ospelt 한국자동차공학회 1998 한국자동차공학회 Workshop Vol.1998 No.-

The prosperity of modem society is unthinkable without technology. One of the most important technical products so far is undoubtedly the automobile.<br/> It has been used by man to satisfy his deep-rooted desire for freedom, independence and mobility. 500 million motor vehicles on the roads across the world are clear evidence of the fact that society no longer wants to forego the benefit of the automobile.<br/> According to the forecasts of several experts, the car population may reach 1.3 billion by the middle of the next century - if the markets keep developing as expected.<br/> This enormous growth rate places an obligation on engineers in the automobile industry to do everything to further improve the automobile.<br/> Above all, this demands an additional deduction of fuel consumption as well as efforts to keep negative impact on the environment to a minimum.<br/> Modem high speed direct injection diesel engines distinguish themselves from other intemal combustion engines through their high efficiency. This efficiency contributes significantly to the management of available crude, oil reserves. Moreover, two further consequences of their low consumption are of increasing interest: Firstly, there is the opportunity for a positive effect on the balance of foreign trade. Secondly, and no less important, there is the opportunity to reduce the atmospheric burden of CO2 emissions,<br/> Given that environmentally responsible, i.e. thrifty and more sustainable, consumption is becoming increasingly important to the human conscience world-wide, the automotive industry's ability to deliver the appropriate technology is currently very much in demand.<br/> A few years ago, only limited improvements in diesel fuel economy seemed plausible. However, in the meantime intensive developments in the fields of fuel injection, supercharging and materials have opened up the potential for fuel consumption improved to a practicable level of 3 litres per 100km (78 miles per US gallon, 94 miles per UK gallon) with light weight cars. Such improvements cannot simply be achieved through engine modification. Future success will depend in how well co-ordinated and focussed the collaborators in vehicle development can work together.<br/> According to recent investigations at AVL, it seems plausible that improvements beyond the previously mentioned fuel consumption level could be achieved though attention to the total drivetrain system, specifically through optimisation of friction parameters, rapid engine warm-up strategies and gear change strategies.<br/> Nevertheless, the conflict between achievement of pollutant emission limits and the minimisation of fuel consumption and noise remains, albeit against a background of the availability to engineers of new and highly advanced development tools.<br/> This paper deals principally with those technologies and methods which offer the best possible fuel economy, whilst still keeping the ever tighter emission and noise standards in mind. And some space<br/> is being reserved for exhaust gas aftertreatment technology. .<br/> The paper would be incomplete without some discussion of the links between engine and vehicle fuel economy.

Comparison of Independent Component Analysis and Blind Source Separation Algorithms

Oh, Sang-Hoon,Cichocki, Andrzej,Choi, Seungjin,Lee, Soo-Young 목원대학교 멀티미디어신기술연구소 2002 멀티미디어신기술연구소논문집 Vol.2 No.1

Various blind source separation (BSS) and independent component analysis (ICA) algorithms have been developed. However, comparison study for recently developed BSS/ICA algorithms has not been extensively carried out yet. The main objective of this paper is to compare various promising BSS/ICA algorithms in terms of several factors such as robustness to sensor noise, computational complexity, the conditioning of the mixing matrix, and the numbers of sources and training samples. We propose several benchmarks which are useful for the evaluation of the algorithms. This comparison study will be useful for real-world applications, especially EEG/MEG analysis and VLSI implementation of these algorithms.

Nonnegative tensor factorization for continuous EEG classification.

Lee, Hyekyoung,Kim, Yong-Deok,Cichocki, Andrzej,Choi, Seungjin World Scientific 2007 International journal of neural systems Vol.17 No.4

<P>In this paper we present a method for continuous EEG classification, where we employ nonnegative tensor factorization (NTF) to determine discriminative spectral features and use the Viterbi algorithm to continuously classify multiple mental tasks. This is an extension of our previous work on the use of nonnegative matrix factorization (NMF) for EEG classification. Numerical experiments with two data sets in BCI competition, confirm the useful behavior of the method for continuous EEG classification.</P>

잡음환경에서 독립성분 분석과 암묵신호분리 알고리즘의 성능비교

오상훈,최승진,이수영,O, Sang-Hun,Cichocki, Andrzej,Choe, Seung-Jin,Lee, Su-Yeong 대한전자공학회 2002 電子工學會論文誌-CI (Computer and Information) Vol.39 No.2

여러 가지의 독립성분분석 및 암묵신호분리 알고리즘들이 개발되었지만, 아직 이러한 알고리즘들의 성능비교가 철저히 이루어지지는 못 하였다. 이 논문은 이 알고리즘들 중에서 뛰어난 알고리즘들을 센서 잡음에 대한 강인성, 계산 복잡도, 혼합 행렬의 조건, 센서 수, 학습패턴 수 등 여러 측면에서 비교한다. 또한, 알고리즘들의 성능 비교에 유용한 문제들도 제시한다. 이 비교결과는 이 알고리즘들의 EEG/MEG 분석, 음성신호분리 등과 같은 실질적 응용에 큰 도움이 될 것이다. Various blind source separation (BSS) and independent component analysis (ICA) algorithms have been developed. However, comparison study for BSS/ICA algorithms has not been extensively carried out yet. The main objective of this paper is to compare various promising BSS/ICA algorithms in terms of several factors such as robustness to sensor noise, computational complexity, the conditioning of the mixing matrix, the number of sensors, and the number of training patterns. We propose several benchmarks which are useful for the evaluation of the algorithm. This comparison study will be useful for real-world applications, especially EEG/MEG analysis and separation of miked speech signals.

연속적인 뇌파 분류를 위한 비음수 텐서 분해

이혜경(Hyekyoung Lee),김용덕(Yong-Deok Kim),Andrzej Cichocki,최승진(Seungjin Choi) 한국정보과학회 2008 정보과학회 컴퓨팅의 실제 논문지 Vol.14 No.5

본 논문에서는 연속적인 뇌파 분류를 위해 비음수 텐서 분해를 이용한 특징 추출과 비터비 알고리즘을 이용한 연속적인 데이타의 클래스 분류를 결합한 새로운 알고리즘을 제시한다. 비음수 텐서 분해는 이미 스펙트럼 데이타에 대해 뇌파의 주요한 특징을 잘 추출한다고 알려진 비음수 행렬 분해의 확장으로써 행렬이라는 제한된 틀에서 벗어나 데이타가 가지는 다양한 차원으로의 확대가 가능하다. 뇌-컴퓨터 인터페이스 컴피티션을 통해 공개된 데이타를 이용한 실험을 통해 제안된 방법의 유용함을 증명하도록 하겠다. In this paper we present a method for continuous EEG classification, where we employ nonnegative tensor factorization (NTF) to determine discriminative spectral features and use the Viterbi algorithm to continuously classify multiple mental tasks. This is an extension of our previous work on the use of nonnegative matrix factorization (NMF) for EEG classification. Numerical experiments with two data sets in BCI competition, confirm the useful behavior of the method for continuous EEG classification.