A three-parameter feature fusion gesture recognition method based on FMCW radar

Abstract

The invention provides a three-parameter feature fusion gesture recognition method based on a FMCW (frequency modulated continuous wave) radar. Firstly, the IF signals of different gestures are acquired by radar, from which the angle parameters, range parameters and Doppler frequency shift parameters are obtained, and the corresponding data sets are constructed. Secondly, the data sets of angle, range and Doppler frequency shift are trained in convolutional neural network, and the eigenvalues are obtained and fused. Finally, the fused eigenvalues are sent into the classifier to get the gestureclass. The invention innovatively proposes a fusion method of multiple parameters in a convolution neural network, solves the problem that the use condition of the traditional recognition algorithm is limited, and the classification effect of the final gesture is better than that of a single parameter.

Description

technical field [0001] The invention relates to the technical field of gesture recognition, in particular to a three-parameter feature fusion gesture recognition method based on FMCW radar. Background technique [0002] With the development of human-computer interaction technology in today's era, gesture recognition has become an important part of human-computer interaction. Its research and development affect the naturalness and flexibility of human-computer interaction, and it has been widely used in various fields. [0003] Traditional human-computer interaction methods rely on input devices such as keyboards and mice, which cannot achieve simple, efficient, and flexible information interaction. Gestures can express information intuitively and effectively, and their information can be conveyed to the visual system. In recent years, due to the popularity of electronic products and the development of related technologies, gesture recognition technology has become a hot res...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a three-parameter feature fusion gesture recognition method based on FMCW radar. Compared with the traditional gesture recognition technology, the present invention combines the distance, Doppler and angle of the gesture action for multi-dimensional parameter learning, effectively solving the problem. Solved the problem of the low amount of gesture description information in the single-dimensional parameter gesture recognition method and the problem that the input samples of the traditional convolutional neural network cannot accommodate multi-dimensional parameters

It can effectively solve the problem of limited use environment of traditional recognition and poor algorithm robustness, and can realize accurate classification of various gestures

Images