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

K-SVD learning dictionary based woven fabric texture flaw detection method

A defect detection and dictionary learning technology, applied in the field of image analysis, can solve problems such as large amount of calculation and unstable detection effect

Active Publication Date: 2018-06-12
DONGHUA UNIV
View PDF7 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the problem of large amount of calculation and unstable detection effect of the woven fabric flaw detection method of the K-SVD learning dictionary in the above-mentioned prior art, and to provide a learning method based on K-SVD that is convenient and fast in calculation and stable in detection effect. A dictionary-based method for detecting flaws in woven fabrics

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
  • K-SVD learning dictionary based woven fabric texture flaw detection method
  • K-SVD learning dictionary based woven fabric texture flaw detection method
  • K-SVD learning dictionary based woven fabric texture flaw detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0082] A woven fabric texture defect detection method based on K-SVD learning dictionary, the specific steps are as follows:

[0083] (1) Image processing;

[0084] First divide the entire woven fabric texture image into n sub-images arranged in rows and columns, and start numbering from 1 to n. The number of each row increases from left to right, and the number of each column increases from top to bottom. Each sub-image has its corresponding number, row number and column number, and then each sub-image is expanded into a column vector, and n column vectors jointly form a test sample image matrix Y, Y=[y 1 ,y 2 ,...y t ,...y n ],y t ∈ R u ,y t The column vector obtained by expanding the sub-image numbered t, t=1,2,...,n,y t =[y lt ,...,y qt ,...,y ut]',y qt for y t The qth element in, q=1,2,...,u, u is y t dimension, the size of the sub-image is 8-64×8-64 pixels;

[0085] (2) Build an initial dictionary D;

[0086] First construct a One-dimensional DCT matrix ...

Embodiment 2

[0138] A kind of woven fabric texture defect detection method based on K-SVD learning dictionary, concrete steps are consistent with embodiment 1, difference is the test sample image selected in step (4), such as Figure 2a As shown, the final defect detection results obtained after the program is completed are as follows Figure 2b Shown, as can be seen from the figure, the detection method of the present invention detects accurately. Run the program code of the embodiment of the present invention 2 many times, select two learning dictionaries y1 and y2 arbitrarily therefrom and implement the regression analysis and obtain the regression model of y=x, i.e. y1=y2, the result is as follows image 3 As shown, it can be seen that the method of the present invention can effectively realize the repeatability of the K-SVD learning dictionary. The results of Examples 1 and 2 show that the learning dictionary algorithm based on K-SVD of the present invention can not only approximate ...

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 a K-SVD learning dictionary based woven fabric texture flaw detection method. The whole woven fabric texture image is decomposed into multiple sub-images, a flaw contained sub-image is obtained by discrimination, and the position of a flaw of a fabric is determined according to the position of the flaw contained sub-image; and flaw discrimination is realized by comparing and reconstructing the sub-images, all the sub-images are unfolded into column vectors and then combined to obtain a test sample image matrix, discrete cosine transform is selected for an initial dictionary, an initial sparse coefficient matrix is solved from the initial dictionary and a training sample image matrix via an orthogonal matching and tracking algorithm, K-SVD dictionary learning is carried out on the training sample image matrix to obtain a dictionary, a sparse coefficient matrix is solved from the dictionary and the test sample image matrix via the orthogonal matching and trackingalgorithm, the test sample image matrix is reconstructed, and column vectors of a reconstructed sample image matrix are converted into reconstructed sub-images. According to the method of the invention, detection is rapid and accurate, and a detection result is stable and highly adaptive.

Description

technical field [0001] The invention belongs to the field of image analysis, and relates to a detection method for texture defects of woven fabrics based on a K-SVD learning dictionary. Background technique [0002] Texture is an important visual feature of object recognition. As an essential problem, texture analysis is widely used in many fields, such as medical diagnosis, product quality inspection and resource remote sensing. The detection of fabric defects can be realized through the texture characterization of fabrics. At present, the detection of fabric defects is mainly based on human visual inspection. Realizing the detection of fabric defects through algorithms can effectively avoid individual errors in human detection on the one hand, and greatly reduce the The productivity is liberated and the labor cost is saved. [0003] Traditional fabric texture characterization methods can be roughly divided into three categories: spectrum-based, statistical-based and model...

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): G06T7/00G06T9/00
CPCG06T7/0004G06T9/00G06T2207/20081G06T2207/30124
Inventor 汪军吴莹史倩倩范居乐江慧肖岚李冠志
Owner DONGHUA UNIV
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