Real-time implementation of electromyography for hand gesture detection using micro accelerometer

Abstract

This paper focuses on the development of a novel approach for identification of various hand movements of a person that involves actions like fist opening, closing, and arm roll. The system consists of an electromyogram (EMG) sensor coupled with a digital MEMS accelerometer (full scale range of ±2 g, ±4 g, ±8 g, and ±16 g) for detection of hand gestures; this system is mounted over a strip strapped on the limb of its user. Based on the analysis of the EMG signals that are coupled with the MEMS accelerometer data from the limb, innumerous hand gestures are identified. Six-point-based calibration of the accelerometer data is done to eliminate mounting errors. The hand movements involving roll are better identified using this sensor topology, which is based on EMG sensor coupled with MEMS accelerometer than a system which just uses an EMG sensor to find out hand gestures because the accelerometer data gives precise information about the orientation of the limb in three-dimensional spaces.