Human body posture recognition method and system based on key point detection technology

Abstract

The invention discloses a human body posture recognition method and system based on a key point detection technology. The method comprises the following steps: (1) acquiring continuous frame images through video data and preprocessing the continuous frame images; (2) carrying out human skeleton key point detection on the acquired image; (3) setting a preset value, determining whether coordinates are converted or not by comparing the number of the detected key points of the human body with the preset value, and carrying out posture recognition; (4) based on the three-dimensional coordinates of the key points, acquiring relative position features of limbs mapped by the key points; (5) calculating a confidence coefficient accumulated weight of the key point, comparing the accumulated weight with a preset threshold value, and judging whether the detected human body key point information representing the posture is sufficient or not; if yes, adopting a limb relative position relation judgment method for posture recognition. According to the invention, posture recognition can be carried out according to the content of the video image.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a human body gesture recognition method and system based on key point detection technology. Background technique [0002] Human body posture detection is one of the most challenging directions in the field of computer vision. It is widely used in security monitoring in key areas, behavior detection in public places, workplace behavior monitoring, driver behavior detection, safety monitoring for the elderly living alone, and security in public security and judicial departments. management and other fields. At present, the detection of human behavior is mostly in the form of on-site supervision by supervisors and video screen monitoring. Due to the large number of monitoring equipment, various types of data, and increasingly complex monitoring objects, the traditional behavior monitoring mode has the following problems: On the one hand, relevant supervision Personnel nee...

AI Technical Summary

Problems solved by technology

At present, the detection of human behavior is mostly in the form of on-site supervision by supervisors and video screen monitoring. Due to the large number of monitoring equipment, various types of data, and increasingly complex monitoring objects, the traditional behavior monitoring mode has the following problems: On the one hand, relevant supervision Personnel need to monitor too much video screen information, insufficient supervision energy, and low work efficiency; on the other hand, there are certain subjective factors in human judgment and supervision, and it is difficult to review and standardize in the later stage

In the prior art, in the process of using the key point technology for gesture recognition and detection, the video content is often directly identified, and the number of key points that can be recognized or whether the characteristic information expressed by the key points is sufficient for discrimination is not considered. The result of gesture recognition is inaccurate or even unable to perform gesture recognition, and the detection efficiency is not high

Images