Abstract
In this paper, we describe a method for recognizing hand motions using a multi-channel surface electromyogram (SEMG), in which the channels are selected using the Monte Carlo method. SEMG measured for the skin surface is widely used as an information source to determine various types of body movements. In our system, we use a 96-channel matrix-type (6×16) surface electrode attached to the forearm in order to measure the SEMG generated from many active muscles during hand movement. In our system, it is very important to select 16 channels that are suitable for recognition from the 96 channels available for the multi-channel electrode. This can be achieved using the Monte Carlo method as follows. The system first generates 1,000 randomly selected 16-channel sets for the multi-channel electrode. These sets are evaluated for the recognition rate of hand movement, and the set of 16 channels that records the highest recognition rate is used for real-time recognition. Using this method, we can select a suitable measurement position of SEMG for each subject. Seven normal subjects were experimentally tested using our system. The recognition rates of 18 hand motions, including 10 finger movements, were assessed for every subject. Using the proposed method, we were able to distinguish all the motions, and the average recognition rate in the real-time recognition experiment was measured to be greater than 97%. We thus conclude that our proposed method will be useful for the recognition of hand motion.
