Preparation of organohydrogel EMG electrodes and their application in dynamic gesture recognition
To address the issue of motion artifacts and poor adherence between traditional electromyographic(EMG)electrodes and the skin,a conductive and self-adhesive organohydrogel was proposed as electrodes for monitoring EMG signals,which was applied to recognize dynamic hand gestures.Firstly,an interpenetrating network structure combining physical and chemical crosslinking was used to prepare the polyacrylamide/sodium alginate/tannic acid organohydroge by adding tannic acid,sodium chloride,and glycerol into a biocompatible polyacrylamide/sodium alginate system.Secondly,the mechanical,conductive and adhesive properties of the prepared organohydrogel were tested.Finally,a convolutional neural network(CNN)algorithm was employed for the recognition of dynamic hand gestures.The results show that the prepared organohydrogel has a low elastic modulus and reversible self-adhesive ability,achieving conformal adhesion to the skin.Meanwhile,the organohydrogel has an elastic modulus of 7.7 kPa,an elongation rate of 310%,an ionic conductivity of 2.81 mS/cm,and good self-adhesive performance as well as degradability.When they are used as skin surface electrodes for EMG signal detection,they can obtain high-quality EMG signals with a signal-to-noise ratio(SNR)of 17.14 dB,and effectively recognize dynamic actions of different interface operations and sign language gestures,with average recognition rates reaching 95.04%and 98.67%,respectively.The organohydrogel electrodes have a high signal-to-noise ratio,can effectively recognize dynamic gestures,and thus have broad application prospects in the fields of human-computer interaction and teleoperation of robots.
国家科技期刊平台
NSTL
万方数据
