Emg Signal Classification Using Fuzzy Logic

Abstract

Electromyography (EMG) signals are an important technique in the control applications of prostatic hand. These signals, which are measured from the skin surface, are used to perform movements such as wrist flexion / extension, forearm supination / pronation and hand opening / closing of prosthetic devices. In this study, root mean square, waveform length and kurtosis methods were applied to extracted EMG signals from flexor carpi radialis and extensor carpi radialis muscles by using two channel surface electrodes. A fuzzy logic based classification method has been applied to classify the extracted signal features. With this method, classification for different gripping movements has been successfully accomplished.