Graph defect recognizing and positioning method based on latent Dirichlet allocation model

A technique of implicit Dirichlet and distribution models, applied in character and pattern recognition, image enhancement, image analysis, etc., can solve the problem of heavy pixel time series, and achieve high recognition rate, accurate recognition and positioning effect

Inactive Publication Date: 2017-12-01
苏州珂锐铁电气科技有限公司
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, identifying a large number of pixel time serie

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
  • Graph defect recognizing and positioning method based on latent Dirichlet allocation model
  • Graph defect recognizing and positioning method based on latent Dirichlet allocation model
  • Graph defect recognizing and positioning method based on latent Dirichlet allocation model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0032] Refer to attached figure 1 to attach Figure 6 , the method for identifying and locating graphic defects based on the hidden Dirichlet distribution model in this embodiment. Before describing the method, first briefly introduce the calculation method of the chaotic eigenvector and the hidden Dirichlet distribution analysis model:

[0033] The literature (F.Taken, Detecting Strange Attractors in Turbulence, Lecture Notes in Mathematics, ed D.A.Rand & L.S.Young, (1981).) points out that the one-dimensional space x(t)=[x 1 (t),x 2 (t),...,x n (t)]∈R n It can be mapped to m-dimensional space, the process is as follows:

[0034]

[0035]

[0036] Here, τ rep...

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 graph defect recognizing and positioning method based on a latent Dirichlet allocation model. The method comprises a training phase and a testing phase. The training phase comprises a step of using the chaos theory to calculate pixel time series of trained images, extracting multiple characteristics, and forming chaotic feature vectors, wherein one trained image is represented by a chaotic feature vector matrix, a step of clustering all trained chaotic feature vector matrixes through a k-means clustering method, forming a code book, and then forming a primary histogram through the code book, and a step of learning the primary histogram through the latent Dirichlet allocation model to obtain advanced histograms, and representing the trained images by the advanced histogram. The testing phase comprises a step of representing a tested image by an advanced histogram 1, a step of calculating a level of similarity between the above advanced histogram 1 and the advanced histograms to judge the defect category and location of the tested image.

Description

technical field [0001] The invention relates to a classification and location method in the technical field of computer pattern recognition, in particular to a pattern defect recognition and location method based on a Latent Dirichlet allocation (LDA) model. Background technique [0002] Traditional material surface defect image modeling methods can be divided into three categories: (1) Based on defect set features (Chen, Y.Q., Nixon, M.S., Thomas, D.W., Statistical Geometric Features for Texture Classification[J].Pattern Recognition,1995,28 (94), pp 537–552.); (2) based on physical characteristics (Reindl, I., O'Leary, P., Geometric Surface Inspection of Raw Milled SteelBlocks[J]. Lecture Notes in Computer Science, 2004, 3212pp 849-856.); (3) Based on statistical features (Habib, M.T., Rokonuzzaman, M., A Set of Geometric Features for NeuralNetwork-Based Textile Defect Classification [J]. International Scholarly Research Notices, 2012, 2012.) [0003] In the existing liter...

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
IPC IPC(8): G06T7/00G06K9/62
CPCG06T7/0002G06T2207/20081G06F18/23213
Inventor 洪金剑王勇
Owner 苏州珂锐铁电气科技有限公司
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