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

Kernel correlation filtering multi-target tracking method fusing motion information

A technology of kernel correlation filtering and multi-target tracking, which is applied in the fields of computer vision and intelligent information processing, and can solve the problems of missed detection, interference, and tracking loss of targets.

Active Publication Date: 2020-06-16
JIANGNAN UNIV
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in multi-target tracking tasks in complex scenes, the detection algorithm is prone to target blurring caused by background clutter interference and violent camera shakes, resulting in missed detection and tracking of targets and false tracking caused by false detection frames. The tracking method still needs to be perfected in many details

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
  • Kernel correlation filtering multi-target tracking method fusing motion information
  • Kernel correlation filtering multi-target tracking method fusing motion information
  • Kernel correlation filtering multi-target tracking method fusing motion information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0081] In order to make the object, technical solution and advantages of the present invention clearer, the implementation manner of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0082] like figure 1 As shown, this embodiment provides an improved kernel correlation filter multi-target tracking method based on motion information, the method comprising:

[0083] Step 1: Initialize the parameters, the parameters include: the target speed of the initial frame (k=1), the target tracking state The total number of video frames N, the total number of video frames N is determined by the number of video frames in the data set, k represents the number of frames in the video, k∈[1,N], the initial frame (k=1 ) is initialized to 0; select the first frame (k=1) with confidence greater than D c The detection box of is used as the initial new target;

[0084] The k frame confidence is greater than D c The number of detecti...

Embodiment 2

[0131] In order to verify the effect of the nuclear correlation filtering multi-target tracking method of fusion motion information described in Example 1, the experiment is as follows:

[0132] 1. Experimental conditions and parameters

[0133] The video training data that the present invention adopts is the sequence 02, 04, 05, 09, 10, 11, 13 these seven groups of video sequences in MOT17, and these seven groups of typical video sequences are all the sequences of multi-target movement under complex scenes, There are surveillance cameras on the street, mobile phone videos of pedestrians, driving recorders on buses, etc., including background clutter interference, close movement of the target, deformation of the target, blurred target, occlusion of the target, frequent and poor movement of the target, camera shake, etc. question. In the experiment, the evaluation algorithm provided by MOTChallengeBenchmark was used, and the evaluation criteria of the algorithm were selected s...

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 kernel correlation filtering multi-target tracking method fusing motion information, and belongs to the field of computer vision and intelligent information processing. On the basis of detection and tracking, the KCF is introduced to track multiple targets, excessive dependence on a detector is reduced, and accurate tracking of the multiple targets is achieved; in the tracking process, speed information and an SCCM mechanism are combined into a tracking framework, so that the problems of tracking of a shielded target and drifting of a tracking frame are solved; and finally, the false target is judged by adopting the IOU and the historical trajectory information, so that trajectory fragments are reduced. Experiments show that the method has good tracking effect androbustness, and can widely meet the actual design requirements of intelligent video monitoring, man-machine interaction, intelligent traffic control and other systems.

Description

technical field [0001] The invention relates to a nuclear correlation filter multi-target tracking method for fused motion information, which belongs to the fields of computer vision and intelligent information processing. Background technique [0002] There are two types of target tracking tasks, single target tracking and multiple target tracking. Single target tracking is to give the size and position of the target frame in the initial frame of the video, so as to realize the precise tracking of the same target in subsequent video frames. At present, the single target tracking algorithm has made great progress with the addition of correlation filtering and deep learning. Correlation filtering is to judge the correlation between two targets by training a filter. In the VOT18 (International Visual Tracking Competition) competition , from the two indicators of accuracy and robustness, more than 50% of the top ten methods use correlation filtering, which shows that correlati...

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): G06T7/246G06K9/62
CPCG06T7/251G06T2207/10016G06T2207/20081G06V10/751G06F18/214G06F18/241
Inventor 杨金龙缪佳妮程小雪李方迪葛洪伟
Owner JIANGNAN UNIV
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