Gesture Recognition Based on sEMG and Support Vector Machine

Abstract

With the continuous development and progress of computer vision, the application scene of human gesture feature recognition is more and more widely, gesture feature recognition in human-computer interaction is a hot topic. The paper proposes a signal-vision model that combines support vector machine and gesture EMG signal to recognize different gesture features and sEMG signals. The model collects different gesture signals through the signal acquisition devices and removes noise through the design of sliding window filter which combines with the machine vision gesture detection framework. The paper also reduces the dimension through SVM models to predict and discriminate two different gesture features and visualizes them, we do the gesture detection and recognition so we can match the different gesture model and sEMG signal acquisition model. We demonstrate the potential of the approach proposed through extensive experiments and results show that the proposed model has significantly distinguish the different gesture features combined with support vector machine which can perform well in recognize and classify different gesture features.