sEMG signal based Human Identification using Hidden Markov Model

초록

Various human identification method has been proposed according to the growing importance of the information security in today's society. Existing methods such as password, PIN (Personal Identification Number) and smart card have been widely used because they are convenient to use and simple. However, these methods have some drawback such as counterfeit, modulation and the risk of information loss. To solve this problem, recently, biometrics, human identification system using biometric information, has been studied. Biometrics is divided into physical characteristics and behavioral characteristics. Biometrics of physical characteristics such as fingerprint, face recognition and iris uses the unique characteristics of the body. Biometrics of behavioral characteristics such as keystroke, signature, voice, eye movement and gait uses the habits of people[1-2]. Among them, the human identification using gait habits includes motion capture, silhouettes and acoustic gait. To acquire the gait data, motion capture uses motion sensors, silhouettes uses cameras and acoustic gait uses sound[3-5]. Recently, sEMG(surface Electromyography) signal has been used to detect gait phase analysis and locomotion mode[2,6]. In this paper, we propose a human identification using HMM(Hidden Markov Model) based on the sEMG signal.