Design and Implementation of Gesture Recognition System Based on Flex Sensors

Sign language is the main means of communication for hearing impaired and mute individuals, special populations, and special situations. However, the current low popularity of sign language has also hindered the deep contact between hearing impaired and mute people and the current society. In order to further promote the application level and service ability of sign language, establish a convenient communication bridge for special populations and scenes, and meet the application requirements of portability, real-time, reliability, and stability, this article designs a gesture recognition system based on multiple sensors. This system combines transfer learning and has functions such as Bluetooth communication, data acquisition and processing, gesture recognition, and result display. The transfer experiment results showed that the accuracy of gesture recognition reached 94.07% and 92.53%, respectively, which were 9.41% and 11.73% higher than before the transfer. Moreover, the accuracy and stability of gesture recognition were improved through transfer learning, and the universality and robustness of the model were improved. Through the testing and gesture recognition experiments of the gesture recognition system, it has been shown that the recognition system can recognize gesture movements of different groups of people, with good recognition rate and stability.