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

A Pedestrian Re-Identification Method Based on Distance Distribution Metric Learning

A technology of pedestrian re-identification and metric learning, which is applied in the field of pedestrian re-identification based on distance distribution metric learning, can solve the problems of complicated calculation, impracticality, and no consideration of distance distribution, and achieves effective solution matrix and optimized fitness function. Effect

Active Publication Date: 2021-10-15
HANGZHOU DIANZI UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method has achieved good experimental results, but the calculation is too complicated to be practical
In fact, the above method only considers the relative distance constraints between the sample pairs, and does not take into account the distribution of the distance itself

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
  • A Pedestrian Re-Identification Method Based on Distance Distribution Metric Learning
  • A Pedestrian Re-Identification Method Based on Distance Distribution Metric Learning
  • A Pedestrian Re-Identification Method Based on Distance Distribution Metric Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] The present invention is further illustrated in conjunction with the accompanying drawings. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. After reading the present invention, those skilled in the art will modify various equivalent forms of the present invention All fall within the scope defined by the appended claims of this application.

[0015] Such as figure 1 As shown, the implementation of the present invention mainly includes four steps: (1) pedestrian image preprocessing, extract image feature, form training database; (2) construct the objective function of learning distance measure; (3) utilize genetic algorithm to obtain optimal Solve the matrix M; (4) Use the obtained matrix to determine the metric function, recognize the tested pedestrian image, and output the recognition result.

[0016] Step 1: Collect images through the camera to build a pe...

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-identification method based on distance distribution metric learning, and belongs to the field of biometric authentication. Firstly, the feature library of pedestrians is constructed, the objective function of learning distance metric is constructed by using Mahalanobis distance, and the distribution characteristics of distance between samples are introduced into the objective function, then the covariance matrix of Mahalanobis distance is solved by genetic algorithm, and finally the obtained The Mahalanobis distance calculates the distance between samples, so as to complete the identification of the pedestrian image and obtain the identification result. The invention focuses on solving the problem of pedestrian re-identification in natural scenes, and aims to improve the recognition rate of the pedestrian re-identification system.

Description

technical field [0001] The invention belongs to the field of biological feature authentication and relates to a pedestrian re-identification method based on distance distribution metric learning. Background technique [0002] Nowadays, surveillance cameras can be seen everywhere in daily life. With the rapid development of computer processing power and parallel computing technology and the increasing demand for video surveillance, pedestrian re-identification technology has emerged as the times require. Pedestrian re-identification refers to pedestrian matching in a multi-camera network under non-overlapping view domains, that is, how to confirm whether the pedestrian targets detected by cameras in different positions at different times are the same person. The practical application value of this technology is very extensive, and it plays a pivotal role in criminal investigation work, image retrieval, and tracing people and relatives. But at the same time, there are also so...

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): G06K9/00G06K9/46G06K9/62G06N3/12
CPCG06N3/126G06V40/103G06V10/56G06F18/22
Inventor 周靖淞甘海涛刘江李明珠刘国攸杨瑞
Owner HANGZHOU DIANZI UNIV