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

Abstract

The present invention specifically relates to a cross-camera pedestrian trajectory processing method, computer equipment and readable storage media. On the basis of yolov4 detection and pedestrian re-identification technology ReID as human figure matching, pedestrian trajectory splicing is performed in combination with computer equipment memory space and hard disk space. When stitching pedestrian trajectories, by judging whether there are trajectories in the preset mapping space, the corresponding pedestrian trajectories are mapped and determined according to different judgment results, and when the pedestrian trajectories are spatially mapped, the mapped Pedestrian trajectories conflict with each other in time or space, which can ensure that there will be no mutual interference between the pedestrian trajectories to be spliced, so as to ensure the accuracy and reliability of trajectory splicing. Further, based on feature similarity Trajectory splicing can take the multi-dimensional characteristics of pedestrians into consideration, so as to completely and accurately determine the global pedestrian trajectory, and avoid the phenomenon of trajectory confusion and trajectory loss.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of pedestrian recognition, and in particular to a cross-camera pedestrian trajectory processing method, computer equipment, and a readable storage medium. Background technique [0002] Person Re-Identification (ReID), also known as cross-mirror tracking, is a popular direction in computer vision research. It mainly solves the recognition and retrieval of pedestrians in cross-camera and cross-scene situations. This technology can recognize pedestrians based on their clothing, body posture, hairstyle and other information. Combining with face recognition, it can be applied to more new application scenarios and raise the cognitive level of artificial intelligence to a new stage. For example, pedestrian re-identification can be applied to intelligent security, intelligent person tracing in large public places, and user trajectory and interest analysis in intelligent shopping malls. However...

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