Target tracking method and system under occlusion state

A target tracking and state technology, applied to the target tracking method and system field in the occluded state, can solve the problems that the similarity of the screen features cannot be greater than the threshold, the tracking target is occluded, and the recognition fails, etc., to achieve the effect of practicability

Active Publication Date: 2020-12-18
重庆特斯联智慧科技股份有限公司
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] For the above functions, it is a difficult problem to track a part or even most of the target area and be blocked by other objects. The target is actually a tracking target, but due to the influence of occluders, the similarity between the extracted picture features and the picture features of the tracking target cannot be greater than the threshold, which can easily cause recognition failure
[0005] The current target tracking technology can only track people and vehicles in the video screen with clear tracking targets, and cannot track partially or even most of the covered tracking targets. However, the blocked tracking target is an inevitable and common phenomenon.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Target tracking method and system under occlusion state
  • Target tracking method and system under occlusion state

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0061] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some of the embodiments of the present invention, but not all of them. Based on the embodiments of the invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0062] Such as figure 1 Shown, the present invention proposes following method:

[0063] A target tracking method in an occlusion state, comprising the steps of:

[0064] S1. Select a tracking target from the first video frame, and extract the picture features of the tracking target in the frame;

[0065] S2. Select one or more candidate targets from the second frame, the third frame, ..., the nth frame of continuous video frames;

[0066] S3, judging whether the candidate target selected by S2 is in an occluded state;

[0067] S4. W...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention proposes a method for tracking a target in an occluded state, the method comprising: S1, selecting a tracking target from the first video frame, and extracting the picture features of the tracking target in the frame; S2, From the 2nd frame, the 3rd frame, ..., one or more candidate targets are selected in the continuous video frame of the nth frame; S3, judge whether the candidate target selected by S2 is in an occluded state; S4, when the candidate target When in the occluded state, use the GAN neural network to de-occlude and reconstruct the candidate target; S5. When the candidate target is in the unoccluded state or after de-occluded reconstruction, perform general target tracking processing on the candidate target to confirm that it is occluded Candidate targets in the picture frames of are the tracking targets. Based on the above method, a matching system is designed. The system uses the GAN neural network to realize the de-occlusion reconstruction of the occluded candidate target, which ensures the normal use of the target tracking technology in the flow of people and vehicles.

Description

technical field [0001] The present invention relates to the technical field of video monitoring, and more specifically relates to a method and system for tracking an object in an occluded state. Background technique [0002] With the continuous expansion of the scale of video surveillance systems, target tracking, as a basic technology in the field of video surveillance, is also constantly developing and progressing, and is used in emergency scenarios such as tracking suspicious persons and tracking illegal vehicles. [0003] Target tracking refers to analyzing and extracting the same target from continuous video frames, such as the same person, vehicle or animal, etc., which belongs to the basic technology in the field of video surveillance. The general method of target tracking is as follows: first, the feature extraction of the tracking target is carried out, that is, the tracking target is selected from the first video frame, and the picture features of the tracking targ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & AuthorityPatents(China)
IPC IPC(8): G06K9/00G06K9/62G06T7/11G06T7/136G06T7/215H04N7/18
CPCG06T7/11G06T7/136G06T7/215H04N7/18G06T2207/20081G06T2207/20084G06V20/42G06F18/22
Inventor王海华马福齐
Owner重庆特斯联智慧科技股份有限公司