采集表面肌电信号应用于动作识别的可行性*

背景：文献表明上肢前臂运动时所产生的表面肌电信号具有非线性特征,而肢体运动时肌电信号又呈现出非平稳特性。目的：设计一种简单的拾取电路采集表面肌电信号,拟应用于动作肌电信号的特征识别。方法：根据表面肌电信号的特点,设计高共模抑制比的前端放大电路,抑制共模干扰;采用低通滤波电路,有源双T带阻滤波器对信号进行去噪处理;对采集得到的信号进行小波包变换,得到信号的特征量。结果与结论：所设计的表面肌电信号检测电路具有较高共模抑制比,并能有效地滤除50 Hz工频信号,可以满足肌电信号采集电路的基本要求。肌电信号的处理结果表明采用子频段能量值的方法可以区分手部4种不同动作。

Feasibility of surface electromyography signal acquisition for action recognition

BACKGROUND: Surface electromyography (sEMG) are widely adopted because of its scot-free. Because of the non-stationary of signals, sEMG signals can be classified in wavelet packet transform to obtain effective parameters.OBJECTIVE: To design a detection circuit according to the characteristics of the sEMG, which can pick-up the SEMG signals for action recognition. METHODS: The high CMMR preamplifier was designed to restrain the common code interference; low-pass filter and active double-T band-stop filter were carried on de-noising processing; sEMG signals could be classified in wavelet packet transform to obtain effective parameters.RESULTS AND CONCLUSION: In the experiment, the circuit could implement the anticipated target, pick-up the sEMG and restrain the interference with 50 Hz; further, four different actions on hands could be recognized by using sub band energy value extracted in the wavelet packet translation of the sEMG.