Target tracking method and device, storage medium

A target tracking and tracking algorithm technology, applied in the information field, can solve problems such as large errors, tracking target loss, tracking failure, etc., and achieve the effects of improving the tracking success rate, reducing the high loss rate, and reducing the amount of tracking calculations

Active Publication Date: 2020-10-16
NINEBOT (BEIJING) TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the process of tracking, tracking is performed based on the information of the current frame image and the previous frame image, but this tracking will accumulate over time, resulting in the accumulation of errors, and eventually the accumulated error is too large, which will cause the tracking target to be blocked. Lost, which leads to the problem of tracking failure

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 device, storage medium
  • Target tracking method and device, storage medium
  • Target tracking method and device, storage medium

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0128] This example is a vision-based multi-target tracking method, which detects multiple candidate targets at regular intervals while tracking. And cluster the candidate targets to determine the state and location of the tracked targets. Finally, the features and clustering results of multiple candidate targets will be stored as the basis for subsequent clustering of candidate targets.

[0129] Such as Figure 4 As shown, the target tracking process of this example can be as follows:

[0130] Selecting a tracking target includes: selecting a tracked target through user interaction, for example, selecting a user facing a camera as a tracking target. Or the user directly selects the tracking target on the mobile phone.

[0131] Pre-set conditions to select the tracking target, such as directly selecting the target with the largest area as the tracking target.

[0132] Short-term tracking: After the target is selected, use methods such as DSST correlation filtering to track...

example 2

[0143] This example can effectively track the current target based on vision for a long time.

[0144] This example combines the target re-detection and multi-target clustering algorithms on the basis of the short-term tracking algorithm. Within an appropriate time interval, target re-detection will be performed to obtain multiple candidate targets, and then the candidate targets will be clustered finally. Determine the characteristics and location of the tracked target. In the long run, this example will continuously confirm the tracked target, correct the position of the tracked target, and ensure the reliability of the tracking result.

[0145] This example uses the Hungarian matching algorithm to transform the clustering problem in the multi-target tracking algorithm into a graph matching problem. The Hungarian matching algorithm can effectively overcome the addition and loss of candidate targets during the tracking process, and can very effectively lock the tracking targ...

example 3

[0148] Such as Figure 6 As shown, this example provides a target tracking method, including:

[0149] Given the information of tracking target A, for example, the position and size of target A, the size here can be understood as the aforementioned scale information;

[0150] Update the information of the current target A;

[0151] Extract the features of the recorded target A on the current image, for example, CNN features;

[0152] input new image;

[0153] Track for a short time, and then return to update the information of the current target A.

[0154] At the same time, it is performed asynchronously every n frames, object detection (obtaining information of multiple objects, for example, information of A, B, or C); here, objects A, B, and C are the aforementioned candidate objects;

[0155] Extract the features of A, B, and C on the image,

[0156] Based on features, match target A (Hungarian matching algorithm);

[0157] Correct or recall the information of target...

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

Embodiments of the present invention disclose a target tracking method and apparatus, and a storage medium. The target tracking method comprises: acquiring a currently collected image; using a first tracking algorithm to perform information processing on the first image information and the first tracking information at a previous moment to obtain the first tracking information at the current moment, wherein the first image information is the image information of the currently collected image; using a second image algorithm to perform information processing on the second image information and the first image feature of the tracking target obtained at a historical moment at predetermined intervals, and determining the second tracking information at the current moment, wherein the second image information is the image information of the currently acquired image, and the historical moment is the moment before the current moment; and according to the second tracking information of the current moment, correcting the tracking information at the current moment.

Description

technical field [0001] The present invention relates to the field of information technology, in particular to a target tracking method and device, and a storage medium. Background technique [0002] The camera collects the image of the tracking target, and then extracts the image information from the collected image to realize the visual tracking of the tracking target. [0003] There are many ways of visual tracking in the prior art, for example, a correlation filtering algorithm is used for target tracking, and these tracking algorithms have the characteristics of a small amount of calculation. However, in the process of tracking, tracking is performed based on the information of the current frame image and the previous frame image, but this tracking will accumulate over time, resulting in the accumulation of errors, and eventually the accumulated error is too large, which will cause the tracking target to be blocked. Lost, which leads to the problem of tracking failure. ...

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 Patents(China)
IPC IPC(8): G06T7/246
CPCG06T7/246G06T2207/10016G06T2207/20084
Inventor 张志敏魏俊生
Owner NINEBOT (BEIJING) TECH CO LTD
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