DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Bien, Zeung-Nam | - |
dc.contributor.advisor | 변증남 | - |
dc.contributor.author | Han, Jeong-Su | - |
dc.contributor.author | 한정수 | - |
dc.date.accessioned | 2011-12-14 | - |
dc.date.available | 2011-12-14 | - |
dc.date.issued | 2006 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=254425&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/36055 | - |
dc.description | 학위논문(박사) - 한국과학기술원 : 전기및전자공학전공, 2006.2, [ xi, 113 p. ] | - |
dc.description.abstract | Recently, bio-signals, such as EMG, EEG, and EOG have been considered as means of communication and control of human-machine systems. The effective recognition of these signals is a promising theme of study since it provides with convenient means for human-machine interaction such as the brain computer interface (BCI) that uses electroencephalogram (EEG), the prostheses controller using electromyogram (EMG), and the mouse using electroolfactogram (EOG). The EMG signal is a form of electric manifestation of neuromuscular activation associated with a contracting muscle. EMG signals have following characteristics. First, EMG signal is nonstationary. Second, EMG signal is different from person to person. There are various factors that affect the physiological bio-signals such as EMG signals. These factors are both internal such as thickness of skins, tissues, and difference of the number of muscle fibers, and external such as electrode-electrolyte interface, and electrode configuration. One of major applications for EMG signals is the control of powered prosthesis. Beyond this rehabilitation application, nowadays, EMG signals are adopted for new human-machine interfaces. A mouse cursor interface, a powered wheelchair interface, and an environment control unit are some representative examples. To apply EMG signals as an input interface to these systems, we must solve the user-dependency problem of EMG signals. To handle this user-dependency problem, we may adopt two very different approaches. One direction is called "personalization" technique. This scheme admits the different characteristics of each user and tries to design personalized pattern recognition structures. To accomplish this task, it may utilize a user-dependent feature selection process and/or an adaptive classifier. In this approach, however, there are two main shortcomings. One is the necessity of an additional system such as a user identification recognition system to make the pattern recognition sy... | eng |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 특징 집합 선택 | - |
dc.subject | Filter Method | - |
dc.subject | 근전도 | - |
dc.subject | 필터방법 | - |
dc.subject | 분리도 | - |
dc.subject | Feature Subset Selection | - |
dc.subject | Separability | - |
dc.subject | EMG | - |
dc.subject | Classifiability | - |
dc.subject | FSS | - |
dc.title | New feature subset selection method and its applications for EMG recognition | - |
dc.title.alternative | 새로운 특징 집합 선택 방법과 근전도 신호 인식에의 응용 | - |
dc.type | Thesis(Ph.D) | - |
dc.identifier.CNRN | 254425/Reference : p. 109-113 | - |
dc.identifier.CNRN | 254425/325007 | - |
dc.description.department | 한국과학기술원 : 전기및전자공학전공, | - |
dc.identifier.uid | 020005341 | - |
dc.contributor.localauthor | Bien, Zeung-Nam | - |
dc.contributor.localauthor | 변증남 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.