Classification of EMG Signal Based on Time Domain and Frequency Domain Features

Abstract

Electromyography (EMG) is widely used in controlling the signal in manipulating the robot assisted rehabilitation. In order to manipulate a more accurate robot assisted, the feature extraction and selection were equally important. This study evaluated the performance of time domain (TD) and frequency domain (FD) features in discriminating EMG signal. To investigate the features performance, the linear discriminate analysis (LDA) was introduced. The present study showed that the FD features achieved the highest accuracy of 91.34% in LDA. The results were verified by LDA classifier and FD features showed best classification performance in EMG signal classification application.