Non-standing posture detection method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a non-standing posture detection method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a pedestrian video and a three-dimensional model associated with the pedestrian video, and carrying out the framing processing of the pedestrian video, and obtaining a plurality of pedestrian images; performing pedestrian area detection on each pedestrian image to obtain pedestrian areas in one-to-one correspondence with the pedestrian images; according to the pedestrian images and the pedestrian areas corresponding to the pedestrian images, constructing a pedestrian trajectory graph in the three-dimensional model; performing shortest path identification on the pedestrian trajectory graph by using a shortest path algorithm to obtain a shortest path trajectory; and performing non-standing posture detection on the shortest path track to obtain a detection result. Therefore, whether the pedestrian video has the non-standing posture segment or not is accurately, scientifically and automatically detected, and the accuracy and efficiency of non-standing posture detection are improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a non-standing posture detection method, device, computer equipment and storage medium. Background technique [0002] The detection of pedestrian standing posture has very important application value in real life, such as sidewalk pedestrian detection, motion detection and medical diagnosis and other fields. At present, the existing method of pedestrian standing posture detection is through a sequence of image data collected by cameras Deep learning achieves the detection result of identifying the standing posture of pedestrians. However, the collected image data will appear in non-standing postures, such as bending over, falling down, etc. These image data will affect the recognition of standing postures and reduce the recognition rate. accuracy, the existing technology usually manually recognizes the image data of the non-standing posture, and then remov...

AI Technical Summary

Problems solved by technology

[0002] The detection of pedestrian standing posture has very important application value in real life, such as sidewalk pedestrian detection, motion detection and medical diagnosis and other fields. At present, the existing method of pedestrian standing posture detection is through a sequence of image data collected by cameras Deep learning achieves the detection result of identifying the standing posture of pedestrians. However, the collected image data will appear in non-standing postures, such as bending over, falling down, etc. These image data will affect the recognition of standing postures and reduce the recognition rate. accuracy, the existing technology usually manually recognizes the image data of the non-standing posture, and then removes the recognized non-standing image data. Therefore, the labor cost is greatly increased, the efficiency is low, and it is not conducive to the pedestrian standing posture identify

Images