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

Pedestrian re-recognition method and device

A pedestrian re-identification and identification technology, applied in the field of image recognition, can solve the problems of loss of spatial information, reduce the accuracy of recognition results, weaken the robustness of pedestrian re-identification, etc., to achieve the effect of improving the effect and improving the accuracy

Pending Publication Date: 2022-01-21
GOSUNCN TECH GRP
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a pedestrian re-identification method and device to solve the problem that the existing pedestrian re-identification method based on deep convolutional neural network loses a lot of spatial information in the process of converting three-dimensional matrix features into one-dimensional features, thus weakening the Problems that increase the robustness of pedestrian re-identification and reduce the accuracy of recognition results

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
  • Pedestrian re-recognition method and device
  • Pedestrian re-recognition method and device
  • Pedestrian re-recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0083] see figure 2 , a pedestrian re-identification device according to an embodiment of the present invention, comprising:

[0084] Obtaining module 1, for obtaining the human body image to be identified;

[0085] The input module 2 is used to input the human body image to be identified into the trained pedestrian re-identification model to obtain the recognition result; wherein, the pedestrian re-identification model includes a backbone network, a multi-channel global generalized average pooling module and a global connection layer;

[0086] The backbone network is used to extract N three-dimensional matrix features of the human body image for identification; wherein, N is greater than or equal to 2;

[0087] The multi-path global generalized average pooling module is used to convert each of the three-dimensional matrix features into one-dimensional features through multi-path global generalized average pooling;

[0088] The fully connected layer is used for:

[0089] ...

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 pedestrian re-recognition method and device. The method comprises the steps of obtaining a to-be-recognized human body image; inputting a to-be-recognized human body image into the trained pedestrian re-recognition model to obtain a recognition result; wherein the pedestrian re-recognition model comprises a backbone network, a multi-channel global generalized average pooling module and a full connection layer; the backbone network is used for extracting N three-dimensional matrix features, used for identity recognition, of the human body image; the multi-channel global generalized average pooling module is used for converting each three-dimensional matrix feature into a one-dimensional feature through multi-channel global generalized average pooling; the full connection layer is used for obtaining a total one-dimensional feature according to each one-dimensional feature; and calculating the similarity between the total one-dimensional feature and the one-dimensional feature of the known identity human body gallery, and taking the identity corresponding to the one-dimensional feature with the highest similarity with the total one-dimensional feature in the known identity human body gallery as a recognition result. According to the embodiment of the invention, the robustness of a pedestrian re-recognition effect is enhanced, and the accuracy of a pedestrian recognition result is improved.

Description

technical field [0001] The present invention relates to the technical field of image recognition, in particular to a pedestrian re-identification method, device and storage medium. Background technique [0002] In indoor places with a high level of confidentiality, only specific personnel can enter and exit a specific area during a specific time period. In response to the confidentiality requirements of this kind of place, access control can be added to specific areas. However, although the addition of access control solves the need for confidentiality of the place, it also has shortcomings. On the one hand, the difficulty and cost of adding access control is directly proportional to the number and subdivision of confidential areas; on the other hand, the existence of access control will bring inconvenience to the passage of staff in confidential places. [0003] In recent years, with the development of computer vision technology, the intelligent analysis and monitoring sy...

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): G06V40/10G06V10/48G06V10/62G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/22
Inventor 刘彪陈利军林焕凯洪曙光刘双广
Owner GOSUNCN TECH GRP
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