A review of Various Classification Algorithms of EMG Signals

Abstract

Electromyography (EMG) signals are muscles signals that enable the identification of human movements without the need of complex human kinematics calculations. Electromyography (EMG) signals are becoming increasingly important in many applications, including clinical/biomedical, prosthesis or rehabilitation devices, human machine interactions, and more. Classification analysis in electromyography (EMG) is very desirable because it allows a more standardized and precise evaluation of the neurophysiological, re-habitational and assistive technological findings. The proper algorithm to classify human movements from raw EMG signals has been an interesting and challenging topic to researchers. Different pattern recognition-based classification algorithms are applied on extracted features to classify the gestures. In current years, many academics have focused on finding suitable features and classifiers to realize high precision. Recent Computational Intelligence studies show that EMG signals can be processed by machine learning methods and more classification analysis in electromyography (EMG) is very desirable because it allows a more standardized and precise evaluation of the neurophysiological, re-habitational and assistive technological findings. The classification accuracy varies according to different classifiers. This paper reviews the common types of classifiers favored by researchers to recognize human movements based on EMG signals. In this paper, Various machine learning methods, like support vector machine (SVM), k-nearest neighbors (KNN), artificial neural network (ANN) and Extreme Learning Machine (ELM) are displayed as classification algorithms. In the present work, the paper has presented an overview of various existing researches in the field of electromyographic signals classification involving various state-of-art techniques