A study on sEMG signals pattern recognition of key hand motions

To improve the living conditions of the amputees, researchers have made various sEMG prosthetic hands. The recognition method of sEMG influences the performance of prosthetic hands greatly. Taking the advantages but mediate the disadvantages of previous studies, this paper puts forward a pattern recognition method to recognize the sEMG signals fast and steadily. This method combines the conventional and emerging control strategy. Three sensors are placed on the forearm to classify seven key motions and one relaxation state. To verify the effect of this method, a series of the experiments are carried out. The obtained sEMG data is analyzed by support vector machine method and neural network method. The experimental results show that the effect of the proposed method is better than that of others, and most of its recognition rates are more than 90%. This proves the feasibility of the method.