A pedestrian re-identification method based on multi-view image feature decomposition

A pedestrian re-identification and image feature technology, applied in character and pattern recognition, instruments, calculations, etc., can solve the problem of low accuracy and achieve the effect of improving accuracy

CN109543602BActive Publication Date: 2020-08-14TAIYUAN UNIV OF TECH
5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2020-08-14

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The present invention relates to the technical field of intelligent image retrieval, and more specifically, relates to a pedestrian re-identification method based on multi-view image feature decomposition. Starting from the target image classification problem, combined with a capsule network, a pedestrian image multi-view image feature generation network is constructed. , decompose any pedestrian image to obtain multi-view image features and the similarity under this view. The features of the same viewing angle are directly used for re-identification, while the features of different viewing angles must undergo feature conversion. In the present invention, the feature conversion matrix is ​​obtained by using the BP neural network, which solves the problem of measuring the features of different viewing angles. Experiments show that the decomposition of pedestrian images is of great help to improve the accuracy of pedestrian re-identification.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present invention relates to the technical field of intelligent image retrieval, and more specifically, to a pedestrian re-identification method based on multi-view image feature decomposition. Background technique

[0002] As social public security is getting more and more attention, a large number of surveillance cameras have been installed in many important places. The surveillance video captured by the surveillance cameras can provide the public security department with clues to detect major criminal cases such as terrorist attacks and mob fights. Pedestrian re-identification is an automatic target recognition technology that quickly locates human targets of interest in the surveillance network, and is an important step in applications such as intelligent video surveillance and human behavior analysis. The main purpose of using pedestrian re-identification technology is to solve the retrieval problem of pedestrians in the field of security moni...

Examples

Embodiment Construction

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0048] A pedestrian re-identification method based on multi-view image feature decomposition is carried out according to the following steps:

[0049] 1. Construct a multi-view image feature decomposition neural network to describe pedestrian images from different perspectives (such as figure 1 shown)

[0050] 1. Establish training data set and test data set of multi-view image feature decomposition neural network

[0051] In the pedestrian image data set, s...