Development of a software module for feature extraction and classification of EMG signals

When the upper limb is amputated or lost, a prosthetic device play an important role in rehabilitation. Prosthetics are available in the form of myoelectric devices. These devices work by sensing the Electromyogram (EMG) signals, through electrodes, when the muscles in the upper arm move, this makes an artificial hand to move accordingly. EMG signals cannot be used directly in raw form to control any prosthetic, it needs to be processed before controlling any device. Different techniques are available for processing the EMG signals. Present paper has used the wavelet transform method to obtain the characteristics or features of EMG signals and then fuzzy controller has been applied on those features so that EMG signals can be used to make prosthetic device functional. NI Lab View, an efficient professional mathematical tool, has been used to process the EMG signals. EMG signals are acquired with the help of sensors placed on the skin and Data Acquisition System. Results of the experiment shows that the developed software module is an effective method to process the EMG signals and can be used for prosthetic devices.