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

Pedestrian re-identification algorithm based on deep learning

A technology of pedestrian re-identification and deep learning, which is applied in biometric recognition, character and pattern recognition, computing, etc., can solve problems such as failure, and achieve the effect of robust pedestrian feature matching ability and high recognition rate

Inactive Publication Date: 2019-08-06
KUNMING UNIV OF SCI & TECH
View PDF16 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The changeable posture of pedestrians leads to the failure of the alignment technology widely used on human faces in ReID

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-identification algorithm based on deep learning
  • Pedestrian re-identification algorithm based on deep learning
  • Pedestrian re-identification algorithm based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: as Figure 1-8 As shown, a pedestrian re-identification algorithm based on deep learning, the specific steps of the method are as follows:

[0027] Step1. Construct a neural network for pedestrian re-identification;

[0028] Step2. Use the parameters trained on ImageNet to assign values ​​to the neural network for pedestrian re-identification;

[0029] Step3. Use the pedestrian re-identification training data set to train the assigned neural network for pedestrian re-identification;

[0030] Step4. Use the trained neural network for pedestrian re-identification to perform pedestrian re-identification on the images in the test data set.

[0031] Further, the parameters trained on ImageNet can be set as follows: the initial learning rate is 0.06, and the learning rate is decayed to 0.012 and 0.0024 at 20 epochs and 40 epochs, respectively.

[0032] Further, the training pictures in the pedestrian re-identification training data set can be set to be random...

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 algorithm based on deep learning. The method comprises the following steps of step 1, constructing a neural network for pedestrian re-identification; step 2, assigning the parameters trained on ImageNet to the neural network for pedestrian re-identification; step 3, using the pedestrian re-identification training data set to train the neural network which is assigned with the value and is used for pedestrian re-identification; and step 4, using the trained neural network for pedestrian re-identification to carry out pedestrian re-identification on the images in the test data set. The method has a more robust pedestrian feature matching capability and can be effectively used for pedestrian recognition, and the recognition rate is higherthan that of an existing known method.

Description

technical field [0001] The invention relates to a pedestrian re-identification algorithm based on deep learning, which belongs to the field of image recognition. Background technique [0002] With the continuous maturity of pedestrian re-identification technology, this technology has also begun to show its great application value. In the field of criminal investigation, the characteristics of the pedestrian re-identification algorithm for modeling human body features are in line with the needs of criminal investigation work for human body image retrieval. In the retail field, pedestrian re-identification algorithms can help supermarket operators to obtain effective customer trajectories and identify customer identities, so as to deeply tap the commercial value that can be used. In smart cities, the successful implementation of corresponding intelligent systems also relies on robust, high-performance pedestrian re-identification algorithms. [0003] The use of convolutional...

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): G06K9/00G06N3/04
CPCG06V40/10G06N3/045
Inventor 李卫疆陈亮雨
Owner KUNMING UNIV OF SCI & TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com