Three-dimensional micro-Doppler gesture identification method based on convolutional neural network

Abstract

The invention discloses a three-dimensional micro-Doppler gesture identification method based on a convolutional neural network and relates to the fields of man-machine interaction, wireless sensing and image processing, in particular to a method for using the convolutional neural network for identifying three-dimensional micro-Doppler gesture temporal-frequency graphs detected by a three-way radar framework. According to the method, firstly, a three-way placement system framework capable of fully collecting gesture speed information is put forward; an energy window statistic technology is used for continuously extracting effective gesture time-domain signals; by means of temporal-frequency graph synthesis, fusion and processing of three-way temporal-frequency graph information can be achieved at the same time. The convolutional neural network which is subjected to cropping and additionally provided with an SVM layer is designed, image information can be fully extracted, and the identification accuracy is high.

Description

technical field [0001] The invention relates to the fields of human-computer interaction, wireless perception and image processing, and in particular to a three-dimensional micro-Doppler gesture time-frequency image recognition method detected by a three-channel radar architecture using a convolutional neural network. Background technique [0002] Gesture, as the most traditional way of human-computer interaction, has been greatly developed, especially when various intelligent miniaturized electronic devices have appeared in people's lives more and more, and the intelligent human-computer interaction of electronic devices has become the focus of everyone's research . At present, gesture recognition technology based on vision and wearable sensors is widely used in the field of human-computer interaction. At present, gesture recognition technology based on vision has been applied to video gesture control and sign language translation recognition of vehicle systems, while gestu...

AI Technical Summary

Problems solved by technology

Foreign researchers use common WiFi signals to realize the recognition of simple gesture signals in the room; and Google has developed a gesture recognition system using high-frequency radar for the operation and control of mobile electronic devices such as micro-shaped smart watches. The position and speed information of the finger realizes the recognition of fine gestures, but the system requires high hardware and costs a lot

Tsinghua University uses support vector machine recognition for the one-dimensional radar micro-Doppler time-frequency map, but the velocity information of gestures is reflected in three dimensions, which leads to insufficient extraction of gesture velocity information and fewer types of gestures

Images