Myoelectric Signal Based Finger Motion Discrimination by using Wavelet’s and Pattern Recognisition

This paper details a strategy of discriminating finger Gestures using surface electromyography (EMG) signals, which could be applied to controlling the advanced multi-fingered myoelectric prosthesis for hand amputees. Finger motions discrimination is the key problem in this study. The EMG signal classification system was established based on the surface EMG signals from the subject’s forearm. Four pairs of electrodes were attached on the subjects to acquire the signals during six types of finger Gestures, i.e. Thumb Extension (TE), Thumb Flexion (TF), Index finger Extension (IE), Index finger Flexion (IF), Middle finger Extension (ME) and Middle finger Flexion (MF). Record electrode signals from the extension digitrum, flexor carpi, extensor policies brevis, flexor digitrum super ficialis are noise filtered and transformed into features using wavelet transforms. Feature sets for six different finger gestures are classified by minimum distance classifier technique. Features construction, recognition accuracy and an approach for an extension of the technique to a variety of real world application areas are presented.