Correlation filtering based end-to-end rapid pedestrian re-identification method

A technology of pedestrian re-identification and correlation filtering, which is applied in character and pattern recognition, instruments, computer components, etc., to achieve the effect of improving accuracy

Inactive Publication Date: 2017-08-22
SUN YAT SEN UNIV
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But in the actual video surveillance environment, the target pedestrians are often hidden in different picture backgrounds and crowds

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
  • Correlation filtering based end-to-end rapid pedestrian re-identification method
  • Correlation filtering based end-to-end rapid pedestrian re-identification method
  • Correlation filtering based end-to-end rapid pedestrian re-identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] Such as Figure 1-5 As shown, the method provided by the invention specifically includes the following steps:

[0024] S1. Extract the color features and gradient direction histogram features of the target pedestrian picture and the picture containing the background and other pedestrians respectively; the color feature and gradient direction histogram feature extracted from the target pedestrian picture are used as positive sample feature data, including background and other pedestrians. The color features and gradient orientation histogram features extracted from pictures of other pedestrians are used as negative sample feature data;

[0025] S2. After performing Fourier transform on the positive and negative sample feature data, train the linear learner as the input training sample to obtain the linear correlation filter matrix of the target pedestrian;

[0026] S3. Scale the monitoring scene pictures at different scales, and then extract color features and gradient ...

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 relates to a correlation filtering based end-to-end rapid pedestrian re-identification method, and overcome the problem that re-identification cannot be implemented unless a lot of candidate pedestrians are marked manually in a present pedestrian re-identification method. Instead, a target pedestrian is directly re-identified from a lot of pictures including pedestrians. End-to-end pedestrian re-identification can be realized needless of pedestrian detection.

Description

technical field [0001] The present invention relates to the technical field of video surveillance, and more specifically, relates to an end-to-end fast pedestrian re-identification method based on correlation filtering. Background technique [0002] Since China's reform and opening up, the national economy has grown rapidly, and the process of urbanization has accelerated in an all-round way. More and more people are pouring into big cities to participate in various constructions of the city. The urban population density is too high, and public safety incidents have attracted much attention. In order to protect the lives and properties of the masses, a large number of surveillance cameras are installed in various public places such as subways, streets and shopping malls. However, in the face of complex monitoring networks and massive monitoring data, it is difficult to achieve rapid processing by relying on human monitoring. Therefore, the need to use computer vision and m...

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): G06K9/00G06K9/46G06K9/66
CPCG06V20/53G06V20/46G06V10/507G06V10/56G06V30/194
Inventor 郑伟诗毛亚芳李伟宏吴伟基
Owner SUN YAT SEN 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
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