Video multi-target tracking method based on multi-Bernoulli characteristic covariance

Abstract

The invention discloses a video multi-target tracking method based on multi-Bernoulli characteristic covariance which belongs to the technical field of artificial intelligent and intelligent information processing and aims mainly to solve the problems that in complex environment, the targets in video multi-target tracking are very close to each other and that the size change and tracking to them are not accurate. This method, according to the multi-Bernoulli filtering framework, is introduced by an integration graph and employs a particle filter method to track video multi-targets whose number are in constant change in combination with the characteristic covariance technology. Based on that, a self-adaptive mechanism and a size self-adaptive mechanism for targets in close contact are proposed to realize the self-adaptive processing of targets in close contact and the tracking windows respectively. Finally, a particle filter method is utilized to realize the self-adaptive recognition and tracking of the video multi-targets through their movement trajectories. The method of the invention is highly robust and has a good anti-interference ability. The method meets the design requirement of a practical engineering system and has good engineering application value.

Description

technical field [0001] The invention belongs to the technical field of intelligent information processing, and relates to a multi-target tracking method of variable number video. Specifically, it is a video multi-target tracking method based on feature covariance matrix and multi-Bernoulli filtering, which can be used for video multi-target detection and tracking in various traffic control, robot navigation and video surveillance systems. Background technique [0002] In the field of computer vision applications, video multi-target tracking with varying numbers of targets, intersecting, close proximity, etc. is a very important and challenging problem, and has always been a hot and difficult point in video tracking research. Especially when the targets intersect or are close to each other, it is easy to cause the target tracking accuracy to drop, or even miss tracking, which directly affects the subsequent target recognition, classification, behavior analysis and other proce...

AI Technical Summary

Problems solved by technology

[0005] For the above problems, the present invention introduces feature covariance technology and particle filter technology under the framework of multi-Bernoulli filtering, and proposes an adaptive variable number video multi-target tracking method to solve the problem that it is difficult to track accurately when the target is close to each other and the size changes. problem, and further introduce particle marking technology on the basis of this method, to realize the accurate tracking of the respective motion trajectories of video objects, and improve the adaptability and robustness of the method of the present invention to video multi-target tracking

Images