Cross-camera pedestrian track processing method, computer equipment and readable storage medium

Abstract

The invention particularly relates to a cross-camera pedestrian trajectory processing method, computer equipment and a readable storage medium, which are used for splicing pedestrian trajectories by combining memory space and hard disk space of the computer equipment on the basis of taking yolov4 detection and pedestrian re-identification technology ReID as human shape matching. When pedestrian trajectory splicing is carried out, whether a trajectory exists in a preset mapping space or not is judged, so that mapping and determination of the corresponding pedestrian trajectory are carried out according to different judgment results, and when spatial mapping is carried out on the pedestrian trajectory, the mapped pedestrian tracks conflict with one another in time or space, the pedestrian tracks to be spliced do not interfere with one another, track splicing accuracy and credibility are guaranteed, further, track splicing is conducted on the basis of feature similarity, multi-dimensionalfeatures of pedestrians can be considered. Therefore, the global pedestrian trajectory can be completely and accurately determined, and the phenomena of trajectory disorder, trajectory loss and the like are avoided.

Description

Technical field [0001] Embodiments of the present invention relate to the field of pedestrian identification, and more particularly to a cross-imaging head pedestrian trajectory, computer device and readable storage medium. Background technique [0002] Person Re-Identification, Reid, also known as cross-mirrors, is the hot direction of computer visual research, mainly to address the identification and retrieval of pedestrians under cross-scenes in cross-scenes. The technology can recognize pedestrians according to information such as pedestrians, and the combination of face recognition, with face recognition, and increase the recognition level of artificial intelligence to a new stage. For example, pedestrian reconstruction can be applied to intelligent security, large public places, intelligent tracing, and analysis of user trajectory interest in smart malls. However, related technologies will have trajectory disorder and trajectory loss when determining the trajectory of globa...

AI Technical Summary

Problems solved by technology

However, when the related technology determines the global pedestrian trajectory, there will be trajectory confusion and trajectory loss.

Images