Target tracking method, computer equipment and storage medium

Abstract

According to a target tracking method, computer equipment and a storage medium, the method comprises the steps of adopting a camera system comprising a plurality of cameras to perform target tracking,each previous frame of image acquired by the plurality of cameras in the camera system is acquired, and an image part where a target object in each previous frame of image is located is used as a reference image; acquiring each current frame image acquired by each camera in the camera system, and acquiring a plurality of alternative image parts preliminarily similar to the corresponding referenceimage in each current frame image to form an alternative image part set of each current frame image; performing target feature extraction on each alternative image part in each alternative image partset to obtain a feature map set corresponding to each current frame image; and processing each feature map set after dimension reduction by using a corresponding filter to obtain a similarity score map of each current frame image, and obtaining an image part with the highest similarity with the corresponding reference image from the similarity score map as a tracking result, so that the accuracyis high.

Description

technical field [0001] The invention relates to the technical field of visual recognition, in particular to a target tracking method, computer equipment and a storage medium. Background technique [0002] In recent years, video surveillance has always been a hot spot in social research, and the use of computer vision technology to process and analyze surveillance video information has become an important development direction in the field of security. Target tracking refers to establishing the positional relationship of the object to be tracked in a continuous video sequence to obtain the complete trajectory of the target. According to the number of cameras used, object tracking can be divided into single-camera tracking and multi-camera tracking. [0003] When tracking a single-camera target, we search around the area predicted for the target position in the previous frame in the current frame, and then find the area most similar to the target in the previous frame as the ...

AI Technical Summary

Problems solved by technology

However, due to the limitation of the field of view of a single camera, the tracking target is prone to occlusion, which affects the tracking result of the frame and is prone to tracking loss.

Moreover, for long-term tracking and monitoring, the trajectory of the target generally crosses cameras, which is also a problem that cannot be solved by single-camera target tracking.

Images