Classification of Finger Movements Using EMG Signals with PSO SVM Algorithm

Abstract

Electromyography (EMG) is a signal produced by human muscles when they contract or relax. This signal is widely used as a controller, for example, to control a robotic arm. This study aims to identify the pattern of finger movement in the form of finger movement using a bracelet-shaped device that has eight EMG sensors. This tool is placed on the lower right hand of a subject to get a signal from the EMG. This study uses the support vector machine (SVM) algorithm combined with the particle swarm optimization (PSO) method. For pattern recognition, the properties of the signal in the time domain are used. From this system, the success of pattern recognition is between 68% to 86%.