A novel approach to recognize hand movements via sEMG patterns

Electromyogram signal (EMG) is an electrical manifestation of contractions of muscles. Surface EMG (sEMG) signal collected form surface of the skin has been used in diverse applications. One of its usages is exploiting it in a pattern recognition system which evaluates and synthesizes hand prosthesis movements. The ability of current prosthesis has been limited in simple opening and closing that decreases the efficacy of these devices in contrary to natural hand. In order to extend the ability and accuracy of prosthesis arm movements and performance, a novel approach for sEMG pattern recognizing system is proposed. In order to have a relevant comparison, present and recent research for designing similar systems was re-evaluated. In this study, we investigate time domain, time-frequency domain and combination of these as a representation of sEMG signal feature for accessing signal information. For pattern recognition of sEMG signals for various hand movements, two intelligent classifiers, namely artificial neural network (ANN) and fuzzy inference system (FIS) were utilized. The results indicate that using compound features with principle component analysis (PCA), dimensionality reduction technique and fuzzy technique for classifier produces the best performance for sEMG pattern recognition system.