A classification method of hand EMG signals based on principal component analysis and artificial neural network

This paper presents a classification method for multi-class classification of electromyography (EMG) signals from eight hand movements. The data were collected from 15 subjects. The EMG signals were extracted using 16 time-domain feature extraction methods. The 16 features are reduced using principal component analysis (PCA) to enhance the classification accuracy. The features results from PCA are classified using artificial neural network (ANN). The classification using ANN result to the training accuracy of 85.7% and the testing accuracy of 81.2%.