Tracking method and device of dynamic target and storage medium

Abstract

The invention discloses a tracking method and device of a dynamic target and a storage medium. The method comprises the steps: acquiring a first frame image from a loaded video, and determining a coordinate position of a target box; selecting a plurality of discrete points from the target box of the first frame image, wherein coordinate positions of the discrete points are first positions, and back-tracking positions of the discrete points in the first frame image as second positions; if distances between the first positions and the second positions are less than a preset value, acquiring a coordinate position of the target box in a second frame image through an optical flow method according to the target box of the first frame image. Compared with the prior art, the tracking method and device of the dynamic target and the storage medium have the advantages that the coordinate position of the target box in the second frame image is determined according to the distances between the first positions and the second positions so as to judge whether tracking is successful so that target tracking accuracy can be effectively improved; Kalman filter prediction helps effectively improve anti-occluding capacity and prevent drifting of a target.

Description

technical field [0001] The invention relates to the field of computer image processing, in particular to a dynamic target tracking method, device and storage medium. Background technique [0002] With the development of computer hardware and image processing technology, dynamic object tracking technology has been widely used in various fields of people's livelihood. For example, in the field of intelligent transportation, it is used to control traffic flow, detect abnormal vehicle behavior, monitor traffic accidents, determine pedestrian behavior, etc. Or in video surveillance, it is used to automatically identify different objects, analyze and extract key useful information in video sources, and quickly and accurately locate At the scene of the accident, judge the abnormal situation in the monitoring screen, and issue an alarm or trigger other actions in the fastest and best way, so as to effectively carry out early warning. Dynamic tracking technology is more widely used ...

AI Technical Summary

Problems solved by technology

However, due to the complex method used in the learning mechanism, the processing time is too long, and the real-time performance needs to be further improved.

In addition, when the moving object encounters occlusion, it is easy to lose the tracking target.

Images