Visual tracking error correction method based on PSR

An error correction and visual tracking technology, applied in the field of computer vision and visual tracking, can solve problems such as sample contamination and incompetence for long-term tracking tasks, and achieve the balance of processing speed, improvement of practicality and use effect, and balance of tracking accuracy

Active Publication Date: 2018-11-02
NANJING UNIV OF POSTS & TELECOMM
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Although the visual target tracking technology has developed rapidly in recent years, the current visual target tracking is only a short-term tracking and cannot be used for long-term tracking tasks.
Therefore, problems such as model drift and sample contamination during the tracking process are still issues that need to be studied.

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
  • Visual tracking error correction method based on PSR
  • Visual tracking error correction method based on PSR
  • Visual tracking error correction method based on PSR

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] As shown in the figure, the present invention discloses a PSR-based visual tracking error correction method.

[0048] Specifically, a PSR-based visual tracking error correction method includes the following steps:

[0049] S1. Calculate the correlation filter H of the first frame using the features and groundtruth values ​​in the target state given in the first frame base .

[0050] S2. Limiting the scope of error correction, that is, limiting the scope of tracking error correction.

[0051] S3. Assuming the moment when the target state is evaluated for the first time as x, calculate its PSR index in the time domain, the calculation formula is

[0052]

[0053] For the video moment belonging to the correction range, assuming k, calculate its PSR index in the time domain, the calculation formula is,

[0054]

[0055] Among them, r is the response value, max represents the maximum value, mean represents the average value, and std represents the standard deviation. ...

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 invention discloses a visual tracking error correction method based on PSR. The method comprises the following steps: step S1, calculating a correlation filter of a first frame; step S2, defininga range for implementing error correction; step S3, calculating PSRk and PSRbass; step S4, comparing the PSRk with the PSRbass; step S5, extracting the characteristics of a target in a new state; andstep S6, repeating the above steps to complete the subsequent detection and correction. When the conditions of object shielding, visual angle change, target deformation, surrounding illumination changes and the like occur, the method disclosed by the invention can judge the state estimation in the follow-up tracking detection process by using the correlation filter obtained by training the first frame, and if the first frame has a higher reliability than a filter which is being used in a current frame, the first frame is used for carrying out state estimation, so that the probability of erroroccurrence in the visual tracking process is reduced, and the precision of visual tracking is improved.

Description

technical field [0001] The invention relates to a visual target tracking adjustment method, in particular to a PSR-based visual tracking error correction method, and belongs to the technical field of computer vision and visual tracking. Background technique [0002] Visual object tracking is a task of great research value in computer vision. Its main applications include unmanned driving, intelligent security, intelligent transportation, etc. However, due to the problems of rotation, occlusion, rapid change, motion blur and complex background of the target appearance during the tracking process, the visual tracking technology is difficult in practical applications. [0003] At present, the commonly used visual target tracking methods are divided into two categories, generative tracking methods and discriminative tracking methods. The previous tracking methods were mainly generative: for example, the ASLS tracker proposed a target tracking method based on the structural loc...

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 & Authority Applications(China)
IPC IPC(8): G06T7/246G06T7/262
CPCG06T7/246G06T7/262G06T2207/20081G06T2207/20056G06T2207/10016
Inventor 王彩玲徐烨超荆晓远
Owner NANJING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap