Finger motion pattern recognition based on sEMG Support Vector Machine

Aiming at solving the problems of non-sufficient recognition type, poor stability and real-time performance, this paper proposed a pattern recognition method based on sEMG and Support Vector Machine(SVM). A forearm electrode array plate was designed to collect sEMG signals through experiments. Parameters of SVM classifier were determined by using SVM theory. Besides, teacher samples were trained by using sEMG signals. Finally, the designed SVM classifier was used to verify the effect of pattern recognition of sEMG signals. The results show that SVM classifier has high accuracy in pattern recognition, which verifies the validity and reliability.