Hand and Wrist Movements Classification Using Surface Electromyogram

Abstract

Myoelectric is a biological signal produced from physiological variations in muscle fibers when they contract and relax. The study of muscle activity through the recording and analysis of myoelectric signals is called electromyography. Electromyography can provide a comprehensive view of the operation and performance of internal muscle groups and cells. The objective of this study is an investigation into hand movement classification using surface electromyography (sEMG) signals. For training and evaluating the proposed model, two data sets from the Ninapro project, a publicly available database for prosthetic hand control, were used: DB5 with low‐cost 16 channels and 200 Hz sampling rate setting and DB10 with 12 channels and 1926 kHz sampling rate setup. First, we divided the EMG data into segments using the windowing technique. These segments were then used to extract a set of time features. Finally, the retrieved feature information was loaded into a simple pattern recognition model: Artificial Neural Network. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.