Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Multi-target tracking algorithm based on feature aggregation

A multi-target tracking and algorithm technology, applied in computing, computer components, instruments, etc., can solve the problems of target identity label confusion, data association failure, similarity, etc.

Pending Publication Date: 2021-05-18
JIANGSU UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In complex scenes, the appearance of similar targets is too similar, or the occlusion between targets and obstacles, targets and targets will cause inaccurate features extracted by the re-identification network and failure of data association, which will lead to confusion of target identity labels

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
  • Multi-target tracking algorithm based on feature aggregation
  • Multi-target tracking algorithm based on feature aggregation
  • Multi-target tracking algorithm based on feature aggregation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0045] like figure 1 A multi-target tracking algorithm based on feature aggregation is shown, including the following steps:

[0046] Step 1. Use the target detection algorithm (specifically, the yolov3 target detection algorithm) to detect the target to be detected, and obtain the target image and the detection frame data set in the detection frame Represents the detection frame data of the i-th target at time t, N is the total number of targets, where, x, y are the coordinates of the center point of the bounding box, and a, h are the aspect ratio and height, respectively.

[0047] Step 2. ...

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 multi-target tracking algorithm based on feature aggregation, and the algorithm comprises the steps: detecting a to-be-detected target through employing a target detection algorithm, and obtaining a target image in a detection frame and a detection frame data set; inputting the target image in the detection frame into a re-identification network, and extracting appearance features of the detection target by using the re-identification network to obtain an appearance feature data set; predicting a tracking trajectory by using a Kalman filtering algorithm; matching the detection target with the tracking trajectory by using an association algorithm; updating the successfully matched tracking trajectory by using Kalman filtering; the multi-target tracking algorithm can accurately extract the appearance characteristics of the detected target, and reduces the phenomenon of identity recognition disorder in the tracking process. And meanwhile, matching failure caused by target shielding and sudden deformation can be effectively inhibited by utilizing a two-stage data association matching mechanism.

Description

technical field [0001] The invention belongs to the field of computer vision, in particular to a multi-target tracking algorithm based on feature aggregation. Background technique [0002] Multi-target tracking technology has been widely used in video surveillance, industrial production, intelligent transportation, military and other fields, and it is also a research hotspot in the field of computer vision today. [0003] Using the "detection and tracking" strategy to achieve target trajectory tracking is the mainstream practice of current technology research. In this approach, the target detector is used to detect the target in each frame image, and the number and state parameters of the target are obtained. Then, the embedded re-identification module is used to extract the appearance information of the detected target, combined with the target's motion The feature information correlates the detected targets in adjacent frames to achieve target matching and tracking. [0...

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): G06K9/00G06K9/62
CPCG06V20/46G06V20/48G06F18/22
Inventor 刘志强任世恒陈林
Owner JIANGSU UNIV
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
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