Human body action classification method and device, terminal equipment and storage medium

Abstract

The invention discloses a human body action classification method, which comprises the following steps of: transmitting electromagnetic waves to a target person by utilizing radar so as to obtain a target echo of the target person; obtaining a target spectrogram based on the target echo; and inputting the target spectrogram into a one-dimensional human body action classifier obtained by training to perform classification so as to obtain a human body action classification result of the target person. The invention further discloses a human body action classification device, terminal equipment and a computer readable storage medium. By means of the human body action classification method, the efficiency of obtaining the human body action classification result by the terminal equipment is improved, and the user experience is good.

Description

technical field [0001] The present invention relates to the technical field of image analysis, in particular to a human action classification method, device, terminal equipment and computer-readable storage medium. Background technique [0002] Human action classification techniques play an important role in various applications, such as security defense and smart environments. The early human action classification technology mainly used optical cameras to collect optical images of human actions, and input the collected optical images into the human action classification model to obtain the classification results of human actions. Although traditional optical cameras have high imaging quality, optical cameras are highly dependent on lighting conditions. When choosing bad weather such as rain, snow and haze, the quality of optical images captured by optical cameras is at a premium. [0003] Therefore, the related technology center proposed a human action classification metho...

AI Technical Summary

Problems solved by technology

[0005] The main purpose of the present invention is to provide a human action classification method, device, terminal equipment and computer-readable storage medium, aiming to solve the existing human action classification method in the prior art, which has high computational complexity and slow computational speed , a technical problem that makes the efficiency of obtaining human action classification results low

Images