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

Infrared image defect identification method of principal component-morphology-watershed edge operator

An infrared image and edge operator technology, applied in image enhancement, image analysis, image data processing, etc., can solve the problems of infrared image noise interference and unclear image boundaries, so as to reduce noise interference and reduce a large number of redundant boundaries. , Improve the effect of defect recognition

Pending Publication Date: 2020-10-27
HEILONGJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Infrared thermal imaging can convert unidentifiable infrared radiation into infrared images visible to the human eye. Due to external interference and the noise of infrared acquisition equipment, the infrared image has large noise interference and unclear image boundaries.

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
  • Infrared image defect identification method of principal component-morphology-watershed edge operator
  • Infrared image defect identification method of principal component-morphology-watershed edge operator

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] A principal component-morphology-watershed edge operator infrared image defect recognition method, the method includes the following steps:

[0040] Step 1: First input the infrared image and create its structural elements, and then decompose the structural elements;

[0041] Step 2: Carry out expansion and erosion operations respectively, then set the number of principal components, and establish a covariance matrix;

[0042] Step 3: Calculate its eigenvalue and corresponding eigenvector, and use the eigenvalue to calculate the cumulative contribution rate;

[0043] Step 4: Determine whether to calculate the outer edge of the defect and the inner edge of the defect according to the cumulative contribution rate;

[0044] Step 5: If the cumulative contribution rate is less than 80%, increase the number of principal components and repeat the above process; if the cumulative contribution rate is ≥80%, reconstruct the morphological gradient matrix, and simultaneously calcu...

Embodiment 2

[0051] According to the principle component-morphology-watershed edge operator infrared image defect recognition method described in embodiment 1, the specific process of the first step is: input the infrared image, create a structural element, and the shape of the structural element has a square, a rectangle And rhombus, etc., and then decompose the structural elements to improve execution efficiency, and perform expansion and erosion processing respectively. The definition is as follows:

[0052] (1)

[0053] (2)

Embodiment 3

[0055] According to the principal component-morphology-infrared image defect recognition method of the watershed edge operator described in embodiment 1 or 2, the specific process of the second step is: set the number of principal components, and standardize at the same time, so that the mean is 0, the standard deviation is 1, the covariance matrix is ​​established, and the eigenvalues ​​are calculated and the corresponding eigenvectors , using the eigenvalues ​​for the cumulative contribution rate The judgment is as follows:

[0056] (3)

[0057] (4)

[0058] The concrete process of described step five is:

[0059] According to the cumulative contribution rate Determine whether to calculate the outer edge of the defect and the inner edge of the defect, if the cumulative contribution rate ≥80%, reconstruct the morphological gradient ng(i,j) and simultaneously calculate the outer edge of the defect and the inner edge of the defect,

[0060] Calculate the defec...

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 an infrared image defect identification method of a principal component-morphology-watershed edge operator. Due to external interference and noise influence of infrared acquisition equipment, infrared image noise interference is large, and image boundaries are not clear. The composition of the invention comprises: firstly, an infrared image is input and structural elementsof the infrared image are created; structural element decomposition is performed, and expansion and corrosion operations are respectively carried out; the method further includes setting number of main components, establishing a covariance matrix, calculating a characteristic value and a corresponding characteristic vector; and using the characteristic value to calculate an accumulated contribution rate, judging whether to calculate a defect outer edge and a defect inner edge according to the accumulated contribution rate, respectively calculating the defect outer edge, the defect inner edge,an output outer edge, an output inner edge, an expansion result graph and a corrosion result graph, carrying out median filtering on the output expansion result graph, and then carrying out binaryzation. The invention is used for the infrared image defect identification method of the principal component-morphology-watershed edge operator.

Description

Technical field: [0001] The invention is applied in the field of infrared thermal wave non-destructive testing and at the same time in the field of image processing, and specifically relates to an infrared image edge recognition method based on principal component-morphology-watershed. Background technique: [0002] Infrared thermal imaging can convert unrecognizable infrared radiation into infrared images visible to the human eye. Due to external interference and the noise of infrared acquisition equipment, the infrared image has large noise interference and unclear image boundaries. Invention content: [0003] The purpose of the present invention is to provide a small 3D printed product defect detection device and detection method excited by a hot water bath. [0004] Above-mentioned purpose realizes by following technical scheme: [0005] A principal component-morphology-watershed edge operator infrared image defect recognition method, the method comprises the followin...

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): G06T7/00G06T7/13G06K9/38G06K9/46G06K9/40G01N25/20
CPCG06T7/0004G06T7/13G01N25/20G06T2207/10048G06T2207/20032G06T2207/20192G06V10/28G06V10/30G06V10/44
Inventor 刘玉波张宏顺张思思许秀秀马向阳唐庆菊
Owner HEILONGJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
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