Video multi-target tracking method based on multi-dimensional feature fusion

A multi-target tracking and multi-dimensional feature technology, applied in the field of video multi-target tracking based on multi-dimensional feature fusion, can solve the problems of many trajectory fragments, target overlap, target detection loss, etc., to achieve high operation efficiency, reduce trajectory fragments, and good application. effect of effect

Active Publication Date: 2020-01-10
杭州云栖智慧视通科技有限公司
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In daily surveillance video, overlapping targets, loss of target detection, and rapid movement of targets caused by mutual occlusion of targets are common; as a result, simple position matching and feature similarity matching cannot solve the problem well, resulting in excessive trajectory fragmentation. Too many, the phenomenon of serial numbers (IDSwitch) in the track is serious

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0045] Embodiment 2 uses Kalman prediction to analyze and predict the location of historical trajectory data, and uses the predicted location to match the location of the target, and restricts the similarity of image content to the matching pair, which further improves the reliability of matching.

[0046] In the third embodiment, the specific content of S2 may also be the superposition of the first and second embodiments, as follows:

[0047] If the track container PathTrack already has the history record information of the detection target Detection, then utilize the Kalman filter to the historical record information, and the position information of the detection target Detection should appear for the current frame (including the center position, aspect ratio and height of the detection target Detection and the The speed information (predictDetection) corresponding to the image coordinates is used for real-time prediction to obtain a prediction rectangular area PreRect. Amon...

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 video multi-target tracking method based on multi-dimensional feature fusion, and relates to the technical field of videos and images. The method comprises the following steps of recording the time information, position information and content characteristics of a detection target; calculating the content feature similarity between the detection target and all track containers, and if the maximum similarity value is greater than a first threshold value, determining that the matching is effective; for the unmatched track container, further calculating the change proportion of the historical speed and the temporary speed of the track, detecting the movement included angle and the width change proportion of the target and the movement direction of the historical track, and if the three are all within a given threshold interval, determining that the matching is effective; and storing the effectively matched detection target in the corresponding track container. According to the method, the matching result is real and reliable, and the phenomenon of serial numbers caused by track fragments, shielding and the like is effectively reduced.

Description

technical field [0001] The invention relates to the technical field of video and image processing, in particular to a video multi-target tracking method based on multi-dimensional feature fusion. Background technique [0002] Multi-target tracking technology is an important part of video analysis technology. This technology takes the input detection target detection image, according to the time sequence of appearance, and obtains the movement direction and predicted position through the target historical trajectory analysis, combined with the content characteristics between the targets. Similarity for matching connection. How to effectively match and connect the target in each video frame with the track target in the historical frame is the key to this technology. At present, in the field of multi-target tracking technology, the position of the target is generally predicted (the Kalman trajectory prediction is a relatively mature option), and then the actual position of the...

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): G06T7/246G06N3/08G06N3/04
CPCG06T7/246G06T2207/10016
Inventor 满庆奎徐晓刚李冠华管慧艳
Owner 杭州云栖智慧视通科技有限公司
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