EMG based classification of hand gestures using PCA and ANFIS

Abstract

This paper presents a comparison study between support vector machine (SVM) and adaptive neuro-fuzzy inference system (ANFIS) classification for electromyography (EMG) signals. The EMG signals were acquired from seven hand common gestures. Sixteen features were extracted and were reduced into three new features set using principal component analysis (PCA). The new features set were divided into two for training and testing. The result of ANFIS classification is 91.43% which is higher than SVM classification that has been conducted in previous study.