Publication

Abstract : This paper presents a novel approach to developing a hand gesture recognition model utilizing Electromyography (EMG) sensors for real-time control of various functions on a mobile device. EMG signals, generated by muscle contractions during hand movements, are captured and processed to identify specific hand gestures. The aim of this research is to design an intuitive and efficient application enabling users to interact with their smartphones seamlessly through hand gestures. The proposed system encompasses data collection from EMG sensors attached to different muscles of the hand, signal processing to extract relevant features, and employing machine learning techniques for gesture classification. The model is trained on a diverse dataset of EMG signals corresponding to various hand movements representing commands such as music playback control, photo capture, and call handling. The trained model is integrated into a smartphone application to facilitate gesture-based control of specified functionalities. Preliminary testing demonstrates promising accuracy and responsiveness, indicating the potential for practical implementation of EMG-based hand gesture recognition in enhancing user experience and accessibility of devices.