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

A complex network image recognition method based on the structure balance theory

A complex network and image recognition technology, applied in the field of image recognition, can solve the problem of weakening the recognition effect, and achieve the effect of improving the operation speed, ensuring the rotation, and strengthening the creativity.

Active Publication Date: 2019-02-15
GUANGDONG UNIV OF TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the above methods are closely related to the position and order of pixels in the image, so the recognition effect will be weakened to varying degrees in the face of image plane rotation invariance, translation invariance, scaling invariance and other characteristics.

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 complex network image recognition method based on the structure balance theory
  • A complex network image recognition method based on the structure balance theory
  • A complex network image recognition method based on the structure balance theory

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] A complex network image recognition method based on structural balance theory, specifically including the following steps:

[0036] Step 1: First read the image to get the image matrix X=[X 1 ,...,X m ]∈R m*n , X is an m-dimensional matrix. Among them, m represents the number of test images, n represents the number of pixels in the image, R represents the real field, and T represents the transpose;

[0037] Step 2: Construct a complex network model based on the results of step 1;

[0038] Step 3: Calculate complex network parameters based on the complex network model established in step 2: node degree, average degree, maximum degree;

[0039]

[0040]

[0041] K k =maxK i

[0042] where K i is the node degree, K μ is the mean degree, K m is the maximum degree;

[0043] Step 4: In order to simplify the calculation and reduce the storage space, delete all nodes whose degree is greater than the average degree;

[0044] Step 5: Perform numerical normalizat...

Embodiment 2

[0058] Suppose a sample image feature parameter v=[α v , β v ], respectively calculate the Euclidean distance of a certain sample image and the image feature parameter of the test group, the grouping of the sample image when the Euclidean distance obtains the minimum value is the grouping of the sample image;

[0059] In summary, the complex network image recognition method based on the structural balance theory provided by the present invention can effectively improve the operation speed and the effectiveness of the nodes by establishing a model through the complex network and using the parameters of the complex network to delete similar small nodes in the image. The result of Hadamard product of constructing the structural balance network of the test image and the structural balance matrix is ​​used as the characteristic parameter of image recognition, which makes the parameters of image recognition simple; the useful features in the image are enhanced more effectively by co...

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 relates to the technical field of image recognition, particularly to a complex network image recognition method based on the structure balance theory. The complex network image recognition method based on the structure balance theory provided by the invention can effectively improve the operation speed and the effectiveness of the nodes by establishing a model of the complex networkand deleting small similarity nodes in the image by utilizing the parameters of the complex network. By constructing the product of the structural equilibrium network and the structural equilibrium matrix of the test image as the characteristic parameters of the image recognition, the parameters of the image recognition are simplified. By constructing the structural balance matrix, the useful feature information in the image can be enhanced more effectively, and the rotation and translation invariance of the image can also be guaranteed. And finally, the method can effectively achieve the image recognition, solves the problems that the traditional complex network image recognition technology in the process of image modeling has larger number of nodes and large amount of calculation, and can not extract feature points better, and has the strong creativity.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a complex network image recognition method based on the structural balance theory. Background technique [0002] Image recognition is one of the key research fields of artificial intelligence. Successfully applied in traffic monitoring, biomedical image recognition, handwriting recognition, automatic driving and other fields. At present, many valuable image recognition methods have been formed, such as region-based methods, texture-based image recognition methods, model-based image recognition methods, K-L-based image recognition methods, geometric feature-based image recognition methods, and edge contour-based image recognition methods. Image recognition methods, etc. However, the above methods are closely related to the position and order of pixels in the image, so their recognition effects are weakened to varying degrees in the face of image plane rotation invarian...

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): G06K9/62
CPCG06F18/29
Inventor 马林妹王银河
Owner GUANGDONG UNIV OF 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