Cusor Controlling Based On Hand Gestures' SEMG Signal

Surface Electromyography (SEMG), as an important biological signal, has been used widely in bionics, biofeedback, sports medicine and rehabilitation engineering. Recently, hand gestures recognition technique based on action SEMG signals has been proposed and developed for controlling myoelectric prostheses, assistive devices and other electronic equipments. In such system, SEMG signals collected from musculature areused for the recognition of hand gestures, which can be interpreted as inputs of human- computer interaction. The work of this paper focuses on the following points:1) The pattern recognition of hand gestures'SEMG signal based on neural network includes SEMG signals pre-processing, hand gesture action activities detection, and feature extraction and classification. A band-pass filter was used for the raw SEMG signal preprocessing to suppress noise. Moving average method was used for the detection of the activities related to hand gesture action. MAV(Mean Absolute Value), ZCR(Zero-Crossing Rate), as well as three order AR model coefficients, were used to extract SEMG signal features, BP(Back Propagrate) neural network and SOFM(Self-Organizing Feature Map) neural network was used as classifiers to recognize hand gestures. The result shows that the higher accuracy was obtained while adopting the two networks as classifiers.2) The designing and realizing of cursor controlling based on the platform of Visual Basic 6.0. The result of the pattern recognition of hand gestures'SEMG signal and The API functions attached to Windows are used to control the cursor.