Dense scene cattle identity recognition method based on attention and local features

A local feature and identification technology, applied in the field of biometrics, can solve problems such as the difficulty of identifying cattle

Pending Publication Date: 2021-04-23
HANGZHOU DIANZI UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The present invention solves the problem of difficult identification of cattle in densely occluded scenes of cattle

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
  • Dense scene cattle identity recognition method based on attention and local features
  • Dense scene cattle identity recognition method based on attention and local features
  • Dense scene cattle identity recognition method based on attention and local features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0053] Based on the above problems, the present invention proposes a network model AL-ResNeXt (Attention-Local-ResNeXt) that can effectively extract the surface features of cattle in the application scenario of dense cattle occlusion. In AL-ResNeXt, the attention module CGL (Channel-Global-Local) that focuses on the local information and global information of the feature map is proposed. CGL not only uses the global features of the feature but also digs out the local information of the feature map.

[0054] In order to solve the problem of information loss caused by network forward propagation, the present invention proposes a space information enhancement module SEM (Space-Enhance-Module). This module can fully capture the important visual cues in the image and enhance the surface features of cattle. The features extracted by the model sho...

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 dense scene cattle identity recognition method based on attention and local features. The CGL module and the spatial information enhancement SEM module are designed on the basis of a deep learning technology on the basis of a cattle crowd dense scene, so that the extracted features have higher robustness when serving as cattle recognition bases. Meanwhile, in order to solve the problem of a strong occlusion scene, a GLR module is designed to search for correlation between local features, and the correlation between the local features is used for representing a cow. The invention provides a dense scene cattle identity recognition method based on attention and local features, can effectively solve the cattle identity recognition problem in a cattle dense scene, and can obtain excellent recognition performance through 1: 1 comparison recognition in a complex scene.

Description

technical field [0001] The invention relates to the field of biological identification, in particular to a method for identifying cows in dense scenes based on attention and local features. Background technique [0002] In recent years, animal husbandry, as the leading industry of agriculture and rural economy, has become the basic industry to ensure the consumption of meat, eggs, milk and animal protein intake of urban and rural residents. Especially since 2012, animal husbandry has become an important measure for poverty alleviation in many western regions and provinces with many ethnic minorities. In the traditional breeding industry, dairy and beef cattle breeding is an important part of the farming industry and occupies an important position in the traditional farming and animal husbandry economic system. Therefore, carrying out individual fine breeding is the main research direction of modern breeding. [0003] With the development of animal husbandry in the directio...

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): G06K9/00G06K9/34G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V40/10G06V10/267G06V10/44G06V10/751G06N3/045G06F18/24G06F18/214
Inventor 郑鹏沈雷李琦
Owner HANGZHOU DIANZI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap