Dynamic Gesture Recognition Method Based on Millimeter-Wave Radar

The millimeter wave radar-based dynamic gesture recognition is the use of electromagnetic waves to obtain the motion characteristics of human dynamic gestures, to achieve automatic recognition of gestures. This paper proposes an frequency modulated continuous wave (FMCW) millimeter wave radar gesture recognition technology, the analysis and processing of radar echo, after the radar signal detection, in a complex environment to extract the gesture and radar distance, speed and other features, to get the radar and target azimuth angle, elevation angle, distance-weighted doppler, azimuth elevation angle covariance, azimuth angle-weighted doppler correlation, elevation angle-weighted doppler correlation. The six features are used to classify the six gestures based on the features using a shallow artificial neural network. It is demonstrated that this method can quickly and effectively identify different categories of gestures and has good generalisation to untrained test sets.