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

Multi-pedestrian tracking method and device based on pedestrian and head detection

A technology for head detection and pedestrian detection, applied in computer parts, character and pattern recognition, instruments, etc., can solve the problem of low tracking accuracy

Pending Publication Date: 2022-02-11
WUHAN UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention proposes a multi-pedestrian tracking method and device based on pedestrian and head detection, which is used to solve or at least partially solve the technical problem of low tracking accuracy existing in the prior art

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-pedestrian tracking method and device based on pedestrian and head detection
  • Multi-pedestrian tracking method and device based on pedestrian and head detection
  • Multi-pedestrian tracking method and device based on pedestrian and head detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] An embodiment of the present invention provides a multi-pedestrian tracking method based on pedestrian and head detection, including:

[0062] S1: Use the pedestrian detector and the head tracker to locate the scene target respectively, and obtain the pedestrian detection result and the head detection result;

[0063] S2: Use the association fusion mode to fuse the head detection frame and the pedestrian detection frame of the same pedestrian to obtain a new pedestrian detection frame set;

[0064] S3: According to the relationship between the pedestrian detection frame and the head detection frame, the pedestrian detection frame in the new pedestrian detection frame set is converted into a head detection frame, and the converted head detection frame forms a new head detection frame set ;

[0065] S4: Use the head tracking method to correlate the data of the head detection frames of the adjacent frames in the new head detection frame set to obtain the head track set; ...

Embodiment 2

[0121] Based on the same inventive concept, this embodiment provides a multi-pedestrian tracking device based on pedestrian and head detection, including:

[0122] The detection module is used to respectively utilize the pedestrian detector and the head tracker to locate the scene target, and obtain the pedestrian detection result and the head detection result;

[0123] The fusion module is used to fuse the head detection frame and the pedestrian detection frame of the same pedestrian to obtain a new pedestrian detection frame set by using the associated fusion mode;

[0124] The first conversion module is used to convert the pedestrian detection frame in the new pedestrian detection frame set into a head detection frame according to the relationship between the pedestrian detection frame and the head detection frame, and the converted head detection frame constitutes a new A collection of head detection frames;

[0125] The association module is used to use the head tracking...

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 provides a multi-pedestrian tracking method and device based on pedestrian and head detection, and the method comprises the following steps: firstly positioning pedestrians and heads in an image through employing a pedestrian detector and a head positioning algorithm, and obtaining a pedestrian detection frame and a head detection frame; secondly, designing an association model, and associating a head detection frame and a pedestrian detection frame of the same pedestrian to obtain a new pedestrian detection frame set; then, converting the new pedestrian detection frame set into a head detection frame, and performing multi-pedestrian tracking in a head tracking mode; and finally, converting the obtained head trajectory into a pedestrian trajectory. According to the method provided by the invention, the problems of missing detection, false detection and the like of the pedestrians caused by shielding in the crowded scene are solved, missing detection of the pedestrian detection frame caused by local shielding can be effectively solved, the multi-pedestrian detection quality in the crowded scene is effectively improved, and the tracking accuracy is improved. The effectiveness of the method provided by the invention is proved in quantitative and qualitative tests on an actual monitoring data set.

Description

technical field [0001] The invention relates to the technical field of monitoring target tracking, in particular to a multi-pedestrian tracking method and device based on pedestrian and head detection. Background technique [0002] The task of multi-target tracking is to give a video and output the trajectories of all the targets appearing in the video. The categories of tracking targets can be pedestrians, vehicles and animals. Pedestrians are non-rigid bodies with deformable properties, and are prone to deformation and other problems during the tracking process. Researching pedestrian tracking is a good target example for target tracking. Secondly, pedestrian tracking has a wide range of applications in monitoring scenarios, and can be applied to multimedia analysis, visual monitoring, and sports analysis. As an intermediate task in the field of computer vision, multi-pedestrian tracking has always been the focus of research scholars. [0003] Detection-based multi-pede...

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): G06V20/52G06V40/16G06V10/25G06V10/774G06V10/75G06K9/62
CPCG06F18/22G06F18/214
Inventor 陈军孙志宏梁超陈金柴笑宇王晓芬叶钰高浩胡皓威
Owner WUHAN UNIV