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

A Pedestrian Re-Identification Method Combining Direct Metrics and Indirect Metrics

A pedestrian re-identification and pedestrian technology, applied in the field of image recognition, can solve the problems of ignoring image recognition information and reducing the recognition effect.

Active Publication Date: 2020-06-16
NINGBO UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The indirect measurement method can make good use of the identification information of other images related to the detection image and the target image to measure the similarity between the two, but it ignores the identification information of the image itself, thus reducing the recognition effect

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 Combining Direct Metrics and Indirect Metrics
  • A Pedestrian Re-Identification Method Combining Direct Metrics and Indirect Metrics
  • A Pedestrian Re-Identification Method Combining Direct Metrics and Indirect Metrics

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0033] A pedestrian re-identification method combining direct measurement and indirect measurement proposed by the present invention, its overall realization block diagram is as follows figure 1 As shown, it includes the following steps:

[0034] ① Select a pedestrian image library that includes a training set and a test set. The training set contains pedestrian images for multiple different pedestrians, and the test set also contains pedestrian images for multiple different pedestrians; where the size of the pedestrian image is M × N pixels.

[0035] Set the total number of pedestrian images included in the training set to num; then according to the num pedestrian images in the training set, the training set is divided into the first image library to be identified and the first target image library, the specific process is: in the training s...

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 combining direct measurement and indirect measurement, which firstly determines a first to-be-recognized image library and a first target image library, and extracts the first and second to-be-identified image libraries and the first and second to-be-identified image libraries. The feature vectors of all pedestrian images in the second target image library; then, based on the first to-be-recognized image library and the first target image library, a weighted fusion model of the direct similarity score and the indirect similarity score is obtained by training, as the final similarity score model; Then use the second to-be-recognized image database and the second target image database to test the trained final similarity score model, according to the difference between each pedestrian image in the second to-be-recognized image database and all pedestrian images in the second target image database The final similarity score is used to identify each pedestrian image in the second to-be-recognized image library; the advantage is that the discriminative information of the two images themselves and the discriminative information of other images related to the two images are fully utilized, thereby greatly improving pedestrians. recognition accuracy.

Description

technical field [0001] The invention relates to an image recognition technology, in particular to a pedestrian re-identification method combining direct measurement and indirect measurement. Background technique [0002] Pedestrian re-identification is the core technology in computer vision. Nowadays, many high-level applications in computer vision rely on accurate pedestrian recognition results, such as target tracking and intelligent video surveillance. However, person re-identification has been one of the most challenging tasks in computer vision due to the problems of illumination changes, viewing angle changes, occlusions, etc. in images. As the basis and challenging task of various vision applications, pedestrian re-identification technology has been widely concerned in the field of computer vision. How to better re-identify pedestrians has been deeply studied by relevant institutions at home and abroad. [0003] The main task of metric learning is to learn a linear o...

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): G06K9/00G06K9/62
CPCG06V40/103G06F18/22G06F18/214
Inventor 蒋桧慧郭立君张荣
Owner NINGBO UNIV