Radar gesture recognition based on lightweight convolutional neural network
Yaoyao Dong, Wei Qu, Pengda Wang, Haohao Jiang, Tianhao Gao, Yanhe Shu · Seventh Asia Pacific Conference on Optics Manufacture (APCOM 2021) · 2022
In order to effectively overcome the limitations of traditional gesture recognition technology, a method of gesture recognition using millimeter wave radar is proposed. First, according to the introduction of the millimeter-wave radar system and the description of the echo model, the millimeter-wave radar is used to collect the measured data; then the average cancellation method is used to suppress the clutter of the measured data, and the joint time-frequency analysis technology is used for effective feature extraction; The extracted features are used as the model input, and a lightweight convolutional neural network model is improved. Its recognition rate is over 96%, and it has good recognition performance.