CN102693536A - Defect region extraction method - Google Patents
Defect region extraction method Download PDFInfo
- Publication number
- CN102693536A CN102693536A CN2012100104726A CN201210010472A CN102693536A CN 102693536 A CN102693536 A CN 102693536A CN 2012100104726 A CN2012100104726 A CN 2012100104726A CN 201210010472 A CN201210010472 A CN 201210010472A CN 102693536 A CN102693536 A CN 102693536A
- Authority
- CN
- China
- Prior art keywords
- defect
- image
- scanning
- point
- defect region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention relates to a defect region extraction method which comprises the following steps: firstly, carrying out line or row gradual scanning on an image; in the process of scanning, if finding that the image grey-level difference of a point is greater than a preset threshold value, determining that the point is a suspicious defect point; carrying out expanding along the scanning direction, and respectively executing a boundary local-search algorithm from the up-and-down direction or the left-and-right direction so as to find out two defect boundaries in the up-and-down direction and the left-and-right direction; and until the boundary points acquired from the up-and-down or left-and-right boundary are coincided, extracting a defect region. According to the method, in the process of extracting a defect, an image is only subjected to scanning operation for one time, and after a first point of the defect is detected, the scanning process is simpler, and detection can be performed automatically tightly along the defect region until the defect region is completely extracted; and the detecting method is rapid in speed, and applicable to the defect region extraction with high requirements on rapid edge-region positioning and instantaneity.
Description
Technical field
The invention belongs to the surface imperfection technical field of quality detection, relate to a kind of defect area method for distilling.
Background technology
Aspect the surface imperfection quality detection technology, generally be that image is handled at present, extract defect area earlier, and then carry out feature extraction, defective is carried out discriminator to defect area.Defect area to the phase one extracts, and traditional defect area method for distilling mainly is to carry out various filtering to image, again the image after the filtering is carried out binaryzation, thereby obtains defect area.Like the radioscopic image defective method for distilling based on the subregion adaptive median filter is the one dimension medium filtering based on sweep trace that carries out different directions according to grey scale change characteristics in the image zones of different; And the length of wave filter can be adjusted automatically along with flaw size; Extract the defective of test specimen quickly and accurately; But the medium filtering processing speed is slower, and also more to flaw size adjustment process consumed time.This method processing speed is too slow, is not suitable for the demand of detection in real time, is easy to generate more noise simultaneously, needs subsequent treatment could further confirm defect area.
Summary of the invention
The purpose of this invention is to provide a kind of defect area method for distilling, to solve the problem that existing method for distilling speed is inappropriate for real-time detection slowly.
For realizing above-mentioned purpose, defect area method for distilling step of the present invention is following:
(1) image is carried out the progressively scanning of column or row;
(2) if the gradation of image value difference of finding a point in the scanning process greater than predetermined threshold value, then this is suspicious defect point;
(3) along direction of scanning expansion, respectively from up and down or left and right directions carry out the border local search algorithm, find out up and down or about two defective borders;
(4) up to up and down or the frontier point adopted of border, the left and right sides overlap, extract defect area.
Further, concurrent up and down two boundary searches when image is carried out column scan, when image is carried out line scanning concurrent about two boundary searches.
Defect area method for distilling of the present invention only carries out the single pass computing to image in extracting the process of defective, and after detecting first of defective; Scanning process is just simpler; Can tightly detect automatically, come out by complete extraction, promptly finish up to defect area along defect area.Therefore can find out that this algorithm speed is fast, and many non-defect areas will be not to be detected, be left in the basket automatically.Compare the detection speed of average raising more than ten times with traditional method, be particularly useful for the demanding occasion of rapid edge zone location and real-time.
Description of drawings
Fig. 1 is the method flow diagram of the embodiment of the invention;
Fig. 2 is the theory diagram of the embodiment of the invention.
Embodiment
The defect area method for distilling is like Fig. 1, shown in 2, and concrete steps are following:
(1) image is pursued column scan with the mode of row;
(2) if the gradation of image value difference of finding a point in the scanning process greater than predetermined threshold value, then this is suspicious defect point;
(3) along direction of scanning expansion, carry out the border local search algorithm from both direction up and down respectively, find out up and down two defective border A1 and A2;
(4) frontier point of adopting up to up-and-down boundary overlaps, and extracts defect area.
Border local search algorithm in the step (3) is a prior art, can equal to be published in February, 2005 " computer utility " the 25th the 2nd phase of volume " border local search algorithm and application " referring to Wu Guifang.
1 order of representation direction of scanning is column scan among Fig. 2, first suspicious defect point that 2 expressions are detected, the suspicious defect area of 3 expressions, the seized image of 4 expressions, the border coincide point of 5 expression A1 and A2.
Because digital picture is two-dimentional, scan image, just need can use the mode of Row Column to the row and progressively scanning of row of image, also can use the mode of Column Row, the image that dual mode is onesize, sweep time is the same with effect.If therefore line scanning, be so concurrent about two boundary searches.Concurrent up and down two boundary searches when promptly image being carried out column scan, when image is carried out line scanning concurrent about two boundary searches.
According to image sequence scanning characteristics, the foundation that finds at first is that the gradation of image value difference does not exceed predetermined threshold value, explains that then this is suspicious defect point, and this suspicious defect area of explanation is about to expanded along the direction of scanning by this some beginning simultaneously.Concrete is defective, and perhaps false defect need utilize mode identification method to confirm after the defect area feature extraction.
Because this method is only carried out the single pass computing to image in extracting the process of defective, and after detecting first of defective; Scanning process is just simpler; Can tightly detect automatically, come out by complete extraction, promptly finish up to defect area along defect area.Therefore can find out that this algorithm speed is fast, and many non-defect areas will be not to be detected, be left in the basket automatically.And based on median filter method, the complicacy of median filter method at first, along with participate in element what and produce many times image operation, also need carry out binaryzation afterwards to image, scan image extracts two-value defective border once more, thereby obtains the zone.Can draw this method improves more than ten times than classic method at least.
It should be noted last that: above embodiment is the non-limiting technical scheme of the present invention in order to explanation only, although with reference to the foregoing description the present invention is specified, those of ordinary skill in the art is to be understood that; Still can make amendment or be equal to replacement the present invention, and not break away from any modification or the local replacement of the spirit and scope of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (2)
1. a defect area method for distilling is characterized in that, the step of this method is following:
(1) image is carried out the progressively scanning of column or row;
(2) if the gradation of image value difference of finding a point in the scanning process greater than predetermined threshold value, then this is suspicious defect point;
(3) along direction of scanning expansion, respectively from up and down or left and right directions carry out the border local search algorithm, find out up and down or about two defective borders;
(4) up to up and down or the frontier point adopted of border, the left and right sides overlap, extract defect area.
2. method according to claim 1 is characterized in that: concurrent up and down two boundary searches when image is carried out column scan, when image is carried out line scanning concurrent about two boundary searches.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210010472.6A CN102693536B (en) | 2012-01-13 | 2012-01-13 | A kind of defect area extracting method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210010472.6A CN102693536B (en) | 2012-01-13 | 2012-01-13 | A kind of defect area extracting method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102693536A true CN102693536A (en) | 2012-09-26 |
CN102693536B CN102693536B (en) | 2015-10-28 |
Family
ID=46858940
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210010472.6A Expired - Fee Related CN102693536B (en) | 2012-01-13 | 2012-01-13 | A kind of defect area extracting method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102693536B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110751619A (en) * | 2019-08-28 | 2020-02-04 | 中国南方电网有限责任公司超高压输电公司广州局 | Insulator defect detection method |
CN113870266A (en) * | 2021-12-03 | 2021-12-31 | 中导光电设备股份有限公司 | Method and system for judging authenticity of line defect based on TFT-LCD |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101256157A (en) * | 2008-03-26 | 2008-09-03 | 广州中国科学院工业技术研究院 | Method and apparatus for testing surface defect |
CN101464418A (en) * | 2007-12-18 | 2009-06-24 | 大日本网屏制造株式会社 | Flaw detection method and apparatus |
US20110069894A1 (en) * | 2009-09-24 | 2011-03-24 | Marie Vans | Image defect detection |
CN102175700A (en) * | 2011-01-20 | 2011-09-07 | 山东大学 | Method for detecting welding seam segmentation and defects of digital X-ray images |
-
2012
- 2012-01-13 CN CN201210010472.6A patent/CN102693536B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101464418A (en) * | 2007-12-18 | 2009-06-24 | 大日本网屏制造株式会社 | Flaw detection method and apparatus |
CN101256157A (en) * | 2008-03-26 | 2008-09-03 | 广州中国科学院工业技术研究院 | Method and apparatus for testing surface defect |
US20110069894A1 (en) * | 2009-09-24 | 2011-03-24 | Marie Vans | Image defect detection |
CN102175700A (en) * | 2011-01-20 | 2011-09-07 | 山东大学 | Method for detecting welding seam segmentation and defects of digital X-ray images |
Non-Patent Citations (2)
Title |
---|
刘蕴辉,刘铁,王权良,罗四维: "基于图像处理的铁轨表面缺陷检测算法", 《计算机工程》, vol. 33, no. 11, 30 June 2007 (2007-06-30) * |
吴贵芳,徐科,徐金梧: "边界局部搜索算法与应用", 《计算机应用》, vol. 25, no. 2, 28 February 2005 (2005-02-28) * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110751619A (en) * | 2019-08-28 | 2020-02-04 | 中国南方电网有限责任公司超高压输电公司广州局 | Insulator defect detection method |
CN113870266A (en) * | 2021-12-03 | 2021-12-31 | 中导光电设备股份有限公司 | Method and system for judging authenticity of line defect based on TFT-LCD |
CN113870266B (en) * | 2021-12-03 | 2022-03-11 | 中导光电设备股份有限公司 | Method and system for judging authenticity of line defect based on TFT-LCD |
Also Published As
Publication number | Publication date |
---|---|
CN102693536B (en) | 2015-10-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109410230B (en) | Improved Canny image edge detection method capable of resisting noise | |
CN106446905B (en) | The extracting method for the face crack texture that percolation algorithm and adaptive Canny algorithm blend | |
Li et al. | A local annular contrast based real-time inspection algorithm for steel bar surface defects | |
CN109727244B (en) | Magnetic shoe surface crack detection method | |
US20090202172A1 (en) | Image Inspection Apparatus, Image Inspection Method and Computer Program | |
CN104318556B (en) | Silicon steel plate surface defect image detection method under oil pollution interference | |
CN105447854A (en) | Small-size glass panel surface defect detection method and small-size glass panel surface defect detection system | |
Qiang et al. | An edge-detection method based on adaptive canny algorithm and iterative segmentation threshold | |
CN104282018A (en) | Method for on-line detection of overall diameter of industrial product based on machine vision | |
CN106780464A (en) | A kind of fabric defect detection method based on improvement Threshold segmentation | |
JP2010268009A5 (en) | ||
CN102737377A (en) | Improved method for extracting sub-pixel edge | |
CN104240252A (en) | Detecting Algorithm for cracks of surface of high-temperature billet of machine vision bar | |
CN103914829A (en) | Method for detecting edge of noisy image | |
CN102693536A (en) | Defect region extraction method | |
WO2013148485A3 (en) | Detection of tissue regions in microscope slide images | |
Hu et al. | Heavy rail surface defects detection based on the morphology of multi-scale and dual-structure elements | |
CN106530292B (en) | A kind of steel strip surface defect image Fast Identification Method based on line scan camera | |
CN103632378B (en) | A kind of multi thresholds edge detection method of game thought of drawing based on even point | |
Al-Hameed | Segmentation of radiographic images of weld defect | |
JP5653370B2 (en) | Method for inspecting cracks in solar cells | |
Wang et al. | Study on SDR extraction of ring weld defects of pipeline | |
Hsu et al. | Vision-based detection of steel billet surface defects via fusion of multiple image features | |
JP2015153528A5 (en) | ||
CN108053414A (en) | A kind of machine vision edge detecting technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151028 Termination date: 20160113 |
|
CF01 | Termination of patent right due to non-payment of annual fee |