Complex Surface Electromyography Signal Gesture Recognition Based on Multipathway Featured Scale Convolutional Neural Network

Surface electromyography-based gesture recognition and prosthetic hand control using deep learning (DL) have become increasingly significant in the field of human-computer interaction. This study aims to enhance the control of prosthetic hands driven by complex gestures, addressing the challenge of low-resolution gesture differentiation caused by the coupling and superposition of surface electromyography signals in DL models. We propose a DL-based framework for the recognition of complex surface electromyography signals, utilizing a multipathway approach to acquire raw surface electromyography signals, process them in the time-frequency domain, and extract features using multiscale convolutional networks. The processed surface electromyography features are then analyzed in parallel to enhance accuracy. This method effectively processes multiple signals concurrently and extracts diverse feature sets. By collecting data from six channels, it achieves an 88.56% recognition rate for 16 complex hand gestures, enabling control of ten distinct prosthetic hand movements. By leveraging multidimensional continuous surface electromyography images, we have developed a feature model that resolves the issues of signal coupling and superposition in multichannel surface electromyography data, allowing for precise control of prosthetic hand movements.