Using Image Processing Techniques and HD-EMG for Upper Limb Prosthesis Gesture Recognition.

In this paper we present the results of using image processing techniques for gesture recognition based on High-Density Electromyography (HD-EMG). Here the instantaneous sample of each EMG channel is represented as a pixel of an image that changes with different movements. In this image, various patterns are recognizable as associated to specific gestures. Experiments were performed to compare the use of image feature extraction by dividing the image into patches (blocks). In this case, the effectiveness for gesture recognition of three different patch sizes and locations were tested and compared. The results show the feasibility of using image processing concepts in order to obtain appropriate features for gesture recognition from HD-EMG, in some cases with advantage when compared to conventional methods.