The research of sEMG movement pattern classification based on multiple fused wavelet function

In this paper, multiple fused features were proposed and three different wavelet basis and the discrete wavelet transform were used to decompose the surface EMG signal into multi-scale according to wavelet transform thought of the multi-resolution analysis in order to solve the non-stationary and nonlinear of surface electromyography(sEMG). The experiments were conducted on five subjects with four lower limb motions. Three feature values, including maximum features of db4 wavelet transform in a certain scale, decomposition coefficient features based on dmey wavelet and singular value features of bior3.1 wavelet transform, were extracted from the original sEMG signals and were analyzed in both single and fusion way. The results demonstrated that the proposed multiple fused features can achieve considerably high classification rates in sEMG motion classification task with BP neural network classifier.