Classification of Electromyographic Hand Gesture Signal using Deep Learning

Abstract

The electromyography (EMG) signal measures the electrical activity of muscles and is often described as a function of amplitude, frequency, and phase over time. These signals are commonly employed in both clinical and biomedical applications. They are used to identify neuromuscular disorders and in other activities such as controlling robots and computers. This proposed study utilizes the CNN to analyse the hand gestures and extract valuable information from these gestures. Consecutive training and testing using images were conducted to evaluate the CNN's performance. The findings demonstrate the effectiveness of the proposed methodology in discerning significant features from complex movements.