A Critical Review on Hand Gesture Recognition using sEMG: Challenges, Application, Process and Techniques

Abstract

Hand gesture recognition systems are gaining popularity these days due to the ease with which humans and machines can communicate. The goal of hand gesture development is to improve interactions between humans and computers for the purpose of transmitting ideas. In a typical HGR systems, the main steps followed are, data collection, pre-processing, feature extraction and classification. For every stage, a significant number of techniques are available with various other sub steps. This study gives an overview of modern hand gesture recognition techniques, its Physiological and Anatomical Background, working and challenges faced by these systems. Moreover, the role of artificial intelligence in optimizing the performance of HGR systems is also delineated in this paper. Also, the precision and accuracy of the HGR approaches gets affected by the complexity and diversity of various hand movements, therefore, the need for implementing AI based ML and DL methods keeps on rising. Keeping this in mind, the performance of various ML algorithms in recognizing the visual and sensor-based hand gestures is investigated. Moreover, the commonly utilized framework in detecting hand gestures has been explored in numerous standard datasets.