Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Target tracking method for sample selectable update mechanism, method for rememorizing effective sample, and distance estimation method

A technology of selective update and target tracking, applied in computing, computer parts, character and pattern recognition, etc., it can solve the problems of tracking algorithm failure and erroneous tracking results, so as to improve adaptability, avoid feature information pollution, and improve tracking effectiveness. sexual effect

Pending Publication Date: 2020-05-22
DALIAN NATIONALITIES UNIVERSITY
View PDF2 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But this method has great limitations, it can only be used for linear motion tracking of the target
[0004] Considering that the impact of the complex environment on target tracking is mainly due to the occurrence of wrong tracking results, but the tracking algorithm will fail only when the number of errors accumulates to a certain extent.

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 for sample selectable update mechanism, method for rememorizing effective sample, and distance estimation method
  • Target tracking method for sample selectable update mechanism, method for rememorizing effective sample, and distance estimation method
  • Target tracking method for sample selectable update mechanism, method for rememorizing effective sample, and distance estimation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] Schematic diagram of tracking results in the case of full target occlusion

[0078] This example is aimed at tracking when the target is fully occluded and then partially occluded. Such as Figure 4 As shown, at frame 7, both the method of the present invention and the original correlation filter tracking method maintain normal target tracking. At the 68th frame, the target is completely blocked by similar-looking obstacles, which leads to an error in the scale of the original correlation filter tracking method. After the target reappears in the 91st frame, the target scale of the original correlation filter tracking method is still in the wrong state, but the method of the present invention can still maintain the correct target tracking because the wrong tracking result is forgotten.

Embodiment 2

[0080] Schematic diagram of the tracking results of the target full occlusion and half occlusion

[0081] This example is aimed at tracking when the target is fully occluded and then partially occluded. Such as Figure 5 As shown, at the 14th frame, both the method of the present invention and the original correlation filter tracking method maintain normal target tracking. Then at the 36th frame, the target is completely blocked by obstacles, and at this time the target scale of the original correlation filter tracking method begins to appear wrong. After the target reappeared in 62 frames, the target appeared half-occluded again. At this time, although the original correlation filtering method can still maintain the tracking state, the target scale has largely failed, and the method of the present invention can also maintain the correct target scale and tracking state. .

Embodiment 3

[0083] Schematic diagram of the tracking results of the continuous full occlusion of the target

[0084] This example is aimed at the tracking of the target being continuously fully occluded. Such as Figure 6 As shown, both the method of the present invention and the original correlation filter tracking method can obtain correct tracking results at the fifth frame. When the target is fully occluded for the first time in the 31st frame, the original correlation filtering method can still track the correct target position, but the scale has a slight failure. Afterwards, the target was fully occluded for the second time in the 67th frame. At this time, the original correlation filter tracking method failed to track, and the wrong target tracking state was still maintained until the 102th frame when the target reappeared. However, the method of the present invention is not affected by erroneous target information generated when the target is occluded, and can still maintain a g...

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 target tracking method of a sample selectable updating mechanism, a method for rememorizing an effective sample, and a distance estimation method, and belongs to the technical field of moving target tracking processing. The invention aims to solve the problem that a wrong tracking result influences a tracking algorithm. The method is characterized by comprising the following steps: 3, analyzing the tracking result through an image feature forgetting method, extracting image features of a target region, comparing the image features with a reference image, and forgetting the tracking result with a large difference; 4, verifying the forgotten tracking result in step 3 by using an energy significant memory method, extracting gradient energy of a forgotten result target area, performing significance analysis, rememorizing the tracking result containing the target into a sample library, and performing a forgetting keeping operation and returning to step 2 or tracking ending on the result free of the tracking target. The methods are suitable for all discrimination model type target tracking methods, can prevent the training set from being polluted by the featureinformation of an occlusion object, and improve the adaptability of the target tracking method to target occlusion.

Description

technical field [0001] The invention belongs to the technical field of moving target tracking processing, and in particular relates to a target tracking method with an optional update mechanism for samples. Background technique [0002] Object tracking technology is an important part of computer vision, and its technology is widely used in autonomous vehicles, mobile robots, and intelligent security systems. Accurate target tracking is conducive to accurately understanding the location of the target, which can not only provide more reliable pedestrian coordinates for autonomous vehicles and assisted driving systems to protect the safety of drivers and pedestrians, but also is an essential requirement for high-tech weapons and GPS systems. important information. In practical applications, complex environments such as target occlusion, target deformation, and light-dark changes often have a great impact on the performance of target tracking algorithms. In general, existing ta...

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): G06K9/00G06K9/62G06K9/46
CPCG06V20/56G06V10/40G06V10/751G06F18/214
Inventor 杨大伟毛琳许烨豪
Owner DALIAN NATIONALITIES UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com