Research on hand gesture recognition based on millimeter wave radar

Abstract

Gesture recognition is an important research direction in the new generation of Human Computer Interaction (HCI) technology, which has a wide range of application prospects in automotive collision avoidance, medical care, and entertainment. Radar-based gesture recognition technology is not affected by site and environmental factors, has the advantages of being independent of light intensity and easy to place. In this paper, we use millimeter wave radar to collect seven defined hand gesture signals, and use signal processing algorithms such as Fourier transform to extract the distance and Doppler information of hand gesture signals according to the radar principle, and accumulate them in the time domain to form a distance-time map and a Doppler-time map, and use the two feature maps as the input of the residual network for recognition and classification, respectively. To improve the gesture recognition accuracy the two feature maps are fused to obtain a new feature fusion map, which is then put into the residual network, and the experimental results show that the recognition accuracy of the seven gestures is 98.58%.