Unlock instant, AI-driven research and patent intelligence for your innovation.

An Adaptive Hierarchical Correlation Multi-Target Tracking Method

A multi-target tracking and self-adaptive technology, applied in instruments, computing, character and pattern recognition, etc., which can solve problems such as follow-up, follow-miss, and target appearance similarity.

Active Publication Date: 2019-05-03
NANJING UNIV OF POSTS & TELECOMM
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to solve the problem of wrong tracking and missing target caused by the similar appearance of the target that is easy to occur in the traditional layered data association technology

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
  • An Adaptive Hierarchical Correlation Multi-Target Tracking Method
  • An Adaptive Hierarchical Correlation Multi-Target Tracking Method
  • An Adaptive Hierarchical Correlation Multi-Target Tracking Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] In order to make the purpose of the present invention, implementation scheme and advantage clearer, the specific implementation of the present invention is described in further detail below, and the concrete process of the present invention is as follows figure 1 shown.

[0042] A robust online hierarchical association method, specifically in accordance with the following steps:

[0043] Step 1. The video of the invention is obtained by a video surveillance system or a collection device of a video sensor network.

[0044] Step 2: Use the detector to detect the video sequence, and obtain the detection result of each frame.

[0045] Step 3. Match the detection results of the previous N frames by means of color similarity to form the initial trajectory of each target in the previous N frames, and define the tracking result of the target i in the current t-th frame as Among them, (x, y) is the center coordinate of the target area, (w, h) is the width and height of the ta...

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 belongs to the technical field of intelligent video monitoring, and proposes an adaptive layered association multi-target tracking method. This method uses incremental linear discriminant analysis (ILDA, Incremental Linear Discriminant Analysis) to find a projection matrix that can separate each target as much as possible, and extracts the feature mean vector of each target, and uses it to define the associated cost of matching the trajectory and the candidate target function to guide the growth of continuous and disconnected trajectories. After the matching of the current frame is completed, the difference between the latest matching result of each trajectory and its feature mean vector is calculated. When the difference is greater than a certain threshold, the matching result is used to update the relevant parameters of the ILDA model, otherwise it is not updated. Compared with the prior art, the present invention can reduce the probability of tracking wrong targets due to similar targets in the process of multi-target tracking.

Description

technical field [0001] The invention belongs to the image analysis technology in the field of image data processing or generation, in particular to an adaptive layered association multi-target tracking method, and belongs to the technical field of intelligent video monitoring. Background technique [0002] Multi-target tracking plays an important role in computer vision systems, and has broad prospects in many fields such as human-computer interaction, military guidance, security monitoring, and intelligent transportation. [0003] In recent years, as the detection results provided by detectors have become more and more reliable, detection-based tracking methods have gradually become a major research direction for multi-object tracking. Among them, the hierarchical data association framework has attracted the attention of many people. It divides data association into two steps. The first step is local association. The traditional method is to associate the detection results ...

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): G06K9/62G06K9/46
CPCG06V10/56G06F18/22
Inventor 胡栋张晶栾筱朋张娇
Owner NANJING UNIV OF POSTS & TELECOMM