Hand movement classification by time domain feature extraction in EMG signals

Prosthetic hands empower amputees by automating control through surface electromyography signals, which are electrical recordings of muscle activity during contraction and relaxation, applying statistical feature analysis, are of great importance in biomedical research and related areas. The aim of this study was to classify six electromyography signals representing different hand movements by extracting statistical features in the time domain. The importance of key steps such as filtering, rectification and segmentation was emphasized to obtain an adequate representation of the data. Through feature extraction, the classification of hand opening and closing states was achieved. This highlights the need for careful feature selection in the classification process.