CN107133962A - A kind of diamond saw blade extracting thermal crack method based on rim detection - Google Patents
A kind of diamond saw blade extracting thermal crack method based on rim detection Download PDFInfo
- Publication number
- CN107133962A CN107133962A CN201710549667.0A CN201710549667A CN107133962A CN 107133962 A CN107133962 A CN 107133962A CN 201710549667 A CN201710549667 A CN 201710549667A CN 107133962 A CN107133962 A CN 107133962A
- Authority
- CN
- China
- Prior art keywords
- saw blade
- diamond saw
- image
- crackle
- psnr
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of diamond saw blade extracting thermal crack method based on rim detection, its method is mainly：The image weighting fusion that two kinds of edge detection operator detections of Roberts and Laplce are obtained, obtains band crackle saw blade edge graph；By mathematical morphology closed operation and two kinds of operators respectively in connection with according still further to same weights Weighted Fusion, obtaining flawless saw blade border template figure；The crackle figure of diamond saw blade will be obtained by calculus of differences with crackle saw blade edge graph and flawless saw blade border template figure.Diamond saw blade crackle can be extracted with this method, method is simple, convenient follow-up to crack length, width, position, the quantitative statisticses of shape and analysis, the problem of factory these data can be used in discovery production simultaneously improves, and has good application prospect.
Description
Technical field
The invention belongs to image processing field, more particularly to a kind of diamond saw blade extracting thermal crack side based on rim detection
Method.
Background technology
In process of production, because production technology level is not high, production technology falls behind and quality of materials diamond saw blade
It can be mingled with micro-crack in reason, the diamond saw blade for causing production.On the one hand the generation of crackle influences product quality, on the other hand
Influence is used, once while micro-crack flase drop can cause major accident in use, therefore the defect to diamond surface crackle is examined
It is necessary to survey with analysis.Crack detection technology based on image procossing is due to economical, conveniently, in real time and be widely applied scope
Become the key technology of crack detection in actual production and life.Because the cutter head of diamond saw blade includes skewness
Diamond dust, cause that on the image shot miscellaneous point and the curve lines similar to crackle can be shown, so rim detection and splitting
Line extracts upper and adds difficulty.Single classical operators can only detect the elementary contour edge of diamond saw blade, it is impossible to be had
The lines and crack information of body, or complex edge is detected, but crackle can not be recognized and separated from complex edge.
Complicated edge extracting problem is converted into simple rim detection problem by technical scheme proposed by the present invention.By two
The improvement of individual edge detection operator Weighted Fusion improves the precision and integrality of detection；Edge detection operator and morphology are closed
Computing combines the method for proposing new drawing template establishment；Crack image is obtained with reference to the thought of difference.Method proposed by the present invention
Simply, extracting thermal crack is complete, facilitates the follow-up statistics and analysis to crackle information data, factory can be according to accurate statistical number
It is found that the problem of in producing, is simultaneously improved, with good application prospect.
The content of the invention
It is a primary object of the present invention to provide a kind of diamond saw blade extracting thermal crack method based on rim detection, the party
Method includes the content of two aspects：The improvement of edge detection operator and the establishment of image template.
For up to above-mentioned purpose, the invention provides following technical scheme：A kind of diamond saw leafing based on rim detection
Line extracting method, specifically includes following steps：
（1）Gray processing processing is carried out to the diamond saw blade coloured image of collection；
（2）To gray level image f（x,y）Rim detection is carried out with Laplace operator and Roberts operators respectively, two sides are obtained
Edge image gL（x,y）And gR（x,y）；
（3）Step（2）Two obtained edge-detected image gL（x,y）And gR（x,y）Fusion is weighted according to fusion rule.
Fusion rule is：h（x,y）=α*gL（x,y）+（1-α）*gR（x,y）, wherein, h（x,y）For the image after fusion, α weighs for weighting
Value, α=0,0.1,0.2 ..., 1；
The determination of best weight value：
Y-PSNR is the important indicator for weighing picture quality after treatment.The calculation formula of Y-PSNR is：
Wherein, k is the digit of sampled point, typically 8；PSNR and MSE refer to Y-PSNR and mean square error；M and n are represented
The size of image is m × n；F (x, y) represents artwork picture point（x,y）Pixel value；H (x, y) represents two kinds of operators when weights are α
Picture point after Weighted Fusion（x,y）Pixel value.
Rectangular coordinate system is set up, abscissa is α, and ordinate is PSNR, take α=0,0.1,0.2 ..., obtained when 1 corresponding
PSNR values, are obtained α and PSNR relation curve by the curve matching to point under rectangular coordinate system, obtain α=0,0.1,
0.2 ..., whole PSNR average value, finds corresponding α values, as best weight value in matched curve when 1.Melted by weighting
Conjunction obtains image h（x,y）, it is used as the diamond saw blade edge image with crackle；
（4）Cavity that can be tiny in filler body with closed operation, connects neighbouring object, the profile of smooth object is removed
The dark-coloured details smaller than structural element.Using 3 × 3 circular configuration elements to gray level image f（x,y）Carry out morphology and close fortune
Calculate, fill crackle, obtain flawless image f1（x,y）；
（5）To flawless image f1（x,y）According to step（2）Handled, obtain flawless edge image m1（x,y）And m2（x,
y）；
（6）To step（5）Two obtained edge-detected image m1（x,y）And m2（x,y）According to step（3）Obtained best weights
Value is merged, and obtains fused images m（x,y）, it is used as the diamond saw blade border template image of flawless；
（7）By step（3）With（6）Obtained band crackle diamond saw blade edge image h（x,y）With the diamond saw blade of flawless
Border template image m（x,y）Carry out difference fusion, obtained image as crack image, then carry out that denoising obtains is most
Whole crack image f2（x,y）.
Compared with prior art, complicated edge extracting problem is converted into simply by technical scheme proposed by the present invention
Rim detection problem.The improvement of two edge detection operator Weighted Fusions is improved to the precision and integrality of detection；By edge
Detective operators and closing operation of mathematical morphology combine the method for proposing new drawing template establishment；Crackle figure is obtained with reference to the thought of difference
Picture.Method proposed by the present invention is simple, strong applicability, and the extraction of crackle facilitates the follow-up statistics to crackle information data and divided
Class, the problem of factory can be found in production according to accurate statistics is simultaneously improved, and has good application prospect.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the present invention, this hair
Bright schematic description and description is used to explain the present invention, does not constitute inappropriate limitation of the present invention.In the accompanying drawings：
Fig. 1 is flow chart of the method for the present invention；
Fig. 2 is the gray-scale map of diamond saw blade；
Fig. 3 is to Fig. 2 Laplace operator edge detection results figures；
Fig. 4 is to Fig. 2 Roberts operator edge detection result figures；
Fig. 5 is to the band crack image after Fig. 3 and Fig. 4 Weighted Fusions；
Fig. 6 is to use Laplace operator edge detection results figure again to the first closed operations of Fig. 2；
Fig. 7 is to use Roberts operator edge detection result figures again to the first closed operations of Fig. 2；
Fig. 8 is with the band crack image after weights Weighted Fusion used in Fig. 5 to Fig. 6 and Fig. 7；
Fig. 9 is the crackle figure finally given after being merged to Fig. 5 and Fig. 8 difference.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with drawings and the specific embodiments, to this
Invention is described in further detail.Described embodiment is only a part of embodiment of the invention, rather than whole implementation
Example.Based on the present embodiment, those of ordinary skill in the art obtained on the premise of creative work is not made it is all its
His embodiment, belongs to the scope of protection of the invention.
As shown in figure 1, being flow chart of the method for the present invention, mainly including gray processing, Roberts rim detections, La Pula
This rim detection, Weighted Fusion, closing operation of mathematical morphology and difference fusion.The present invention program specifically includes following steps：
（1）Gray processing processing is carried out to the diamond saw blade coloured image of collection；
（2）To gray level image f（x,y）Rim detection is carried out with Laplace operator and Roberts operators respectively, two sides are obtained
Edge image gL（x,y）And gR（x,y）；
（3）Step（2）Two obtained edge-detected image gL（x,y）And gR（x,y）Fusion is weighted according to fusion rule.
Fusion rule is：h（x,y）=α*gL（x,y）+（1-α）*gR（x,y）, wherein, h（x,y）For the image after fusion, α is for weighting
Number, α=0,0.1,0.2 ..., 1；
Determine best weight value：
Y-PSNR is the important indicator for weighing picture quality after treatment.The calculation formula of Y-PSNR is：
Wherein, k is the digit of sampled point, typically 8；PSNR and MSE refer to Y-PSNR and mean square error；M and n are represented
The size of image is m*n；F (x, y) represents artwork picture point（x,y）Pixel value；H (x, y) represents that two kinds of operators add when weights are α
Picture point after power fusion（x,y）Pixel value.
Rectangular coordinate system is set up, abscissa is α, and ordinate is PSNR, take α=0,0.1,0.2 ..., obtained when 1 corresponding
PSNR values, are obtained α and PSNR relation curve by the curve matching to point under rectangular coordinate system, obtain α=0,0.1,
0.2 ..., whole PSNR average value, finds corresponding α values, as best weight value in matched curve when 1.Melted by weighting
Conjunction obtains image h（x,y）, it is used as the diamond saw blade edge image with crackle；
（4）Cavity that can be tiny in filler body with closed operation, connects neighbouring object, the profile of smooth object is removed
The dark-coloured details smaller than structural element.Using 3 × 3 circular configuration elements to gray level image f（x,y）Carry out morphology and close fortune
Calculate, fill crackle, obtain flawless image f1（x,y）；
（5）To flawless image f1（x,y）According to step（2）Handled, obtain flawless edge image m1（x,y）And m2
（x,y）；
（6）To step（5）Two obtained edge-detected image m1（x,y）And m2（x,y）According to step（3）Obtained best weights
Value is merged, and obtains fused images m（x,y）, it is used as the diamond saw blade border template image of flawless；
（7）By step（3）With（6）Obtained band crackle diamond saw blade edge image h（x,y）With the diamond saw blade of flawless
Border template image m（x,y）Carry out difference fusion, obtained image as crack image, then carry out that denoising obtains is most
Whole crack image f2（x,y）.
To illustrate, carried out with an example as described below.Specific implementation process is：
1st, experimental subjects is the coloured image of collection acquisition diamond saw blade under diamond saw blade coloured image, black background（Figure
Picture size is 389 × 389, and resolution ratio is 96dpi）The image carried out after gray proces is as shown in Figure 2；
2nd, rim detection is carried out to gray level image with two kinds of operators, wherein, gray level image is detected with Laplace operator
Obtained edge image with Roberts operators to gray level image as shown in figure 3, carried out detecting obtained edge image such as Fig. 4 institutes
Show；
3rd, according to step（3）Fusion rule fusion is weighted to Fig. 3 and Fig. 4, obtain weights α=0,0.1,0.2 ..., when 1
Corresponding PSNR values, set up rectangular coordinate system, and abscissa is α, and ordinate is PSNR, and α and PSNR are obtained by curve matching
Corresponding relation, the average value for trying to achieve PSNR is 29.48dB, and the value that corresponding α is found out on curve is 0.7, as best weights
Value, according to best weight value Weighted Fusion Fig. 3 and Fig. 4, obtains the diamond saw blade edge image with crackle as shown in Figure 5；
4th, closing operation of mathematical morphology filling crackle is first carried out to gray-scale map Fig. 2, closed operation uses 3 × 3 circular configuration elements, then divides
Rim detection is not carried out to gray level image Fig. 2 with two kinds of operators, wherein, first with after closed operation with Laplace operator to gray-scale map
As Fig. 2 carries out detecting obtained edge image as shown in fig. 6, first with entering gray level image Fig. 2 with Roberts operators after closed operation
The edge image that row detection is obtained is as shown in Figure 7；
5th, according to step（3）Fusion rule, best weight value is 0.7, and fusion is weighted to Fig. 6 and Fig. 7, the image after fusion
For the diamond saw blade border template image of flawless, as shown in Figure 8；
6th, what is Fig. 5 and Fig. 8 progress difference fusions obtained is exactly crack image, as shown in Figure 9.
In summary, using the extracting thermal crack method of the present invention, diamond saw blade crackle can be extracted, facilitates follow-up counterincision
Line length, width, position, the quantitative statisticses of shape and analysis.Technical scheme proposed by the present invention asks complicated edge extracting
Topic is converted into simple rim detection problem, solves the difficult point that crackle is recognized and separated from diamond saw blade complex edge,
Method is simple, strong applicability, and the extraction of crackle facilitates the follow-up statistics to crackle information data and classification, and factory can be according to essence
The problem of true statistics is found in production is simultaneously improved, and has good application prospect.
Claims (2)
1. a kind of diamond saw blade extracting thermal crack method based on rim detection, it is characterised in that：With Roberts and Laplce
Two kinds of edge detection operators carry out rim detection to diamond saw blade gray-scale map respectively, detect that two obtained images are weighted
Fusion, weights take best weight value, obtain band crackle diamond saw blade edge image；With the closed operation of mathematical morphology to diamond
Saw blade gray-scale map is handled, then with two kinds of edge detection operators respectively in connection with according still further to same best weight value Weighted Fusion, inspection
Measure flawless diamond saw blade border template image；Will be with crackle diamond saw blade edge image and flawless diamond saw
Piece border template image obtains the crack image of diamond saw blade by calculus of differences.
2. a kind of diamond saw blade extracting thermal crack method based on rim detection according to claim 1, it is characterised in that institute
Stating the determination method of best weight value is：Rectangular coordinate system is set up, abscissa is weights, and ordinate is PSNR, and weighting value is 0,
0.1,0.2 ..., corresponding PSNR values are obtained when 1, under rectangular coordinate system by the curve matching to point obtain weights and
PSNR relation curve, obtains weights for 0,0.1,0.2 ..., whole PSNR average value when 1, is found in matched curve pair
The weights answered, as best weight value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710549667.0A CN107133962A (en) | 2017-07-07 | 2017-07-07 | A kind of diamond saw blade extracting thermal crack method based on rim detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710549667.0A CN107133962A (en) | 2017-07-07 | 2017-07-07 | A kind of diamond saw blade extracting thermal crack method based on rim detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107133962A true CN107133962A (en) | 2017-09-05 |
Family
ID=59737409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710549667.0A Pending CN107133962A (en) | 2017-07-07 | 2017-07-07 | A kind of diamond saw blade extracting thermal crack method based on rim detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107133962A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108051445A (en) * | 2017-12-05 | 2018-05-18 | 广东钜鑫新材料科技股份有限公司 | A kind of method for detecting diamond compact underbead crack position |
| CN108765436A (en) * | 2018-04-21 | 2018-11-06 | 卞家福 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
| CN108765435A (en) * | 2018-04-21 | 2018-11-06 | 卞家福 | A kind of optical disc apparatus based on improvement Roberts operators |
| CN110836706A (en) * | 2018-08-15 | 2020-02-25 | 珠海格力电器股份有限公司 | Liquid level detection method and device |
| CN111696067A (en) * | 2020-06-16 | 2020-09-22 | 桂林电子科技大学 | Gem image fusion method based on image fusion system |
| CN111915633A (en) * | 2020-07-24 | 2020-11-10 | 安徽工程大学 | Casting edge detection method and device |
| CN115984316A (en) * | 2023-03-20 | 2023-04-18 | 东莞市兆丰精密仪器有限公司 | Industrial image edge extraction method and device in complex environment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246593A (en) * | 2008-03-27 | 2008-08-20 | 北京中星微电子有限公司 | Color image edge detection method and apparatus |
-
2017
- 2017-07-07 CN CN201710549667.0A patent/CN107133962A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101246593A (en) * | 2008-03-27 | 2008-08-20 | 北京中星微电子有限公司 | Color image edge detection method and apparatus |
Non-Patent Citations (2)
| Title |
|---|
| 任帅: "基于路面检测车图像的裂缝提取方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 赵静 等: "融合Canny算子和小波变换的边缘检测方法", 《计算机仿真》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108051445A (en) * | 2017-12-05 | 2018-05-18 | 广东钜鑫新材料科技股份有限公司 | A kind of method for detecting diamond compact underbead crack position |
| CN108051445B (en) * | 2017-12-05 | 2020-04-28 | 广东钜鑫新材料科技股份有限公司 | Method for detecting crack position in diamond compact |
| CN108765436A (en) * | 2018-04-21 | 2018-11-06 | 卞家福 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
| CN108765435A (en) * | 2018-04-21 | 2018-11-06 | 卞家福 | A kind of optical disc apparatus based on improvement Roberts operators |
| CN110836706A (en) * | 2018-08-15 | 2020-02-25 | 珠海格力电器股份有限公司 | Liquid level detection method and device |
| CN111696067A (en) * | 2020-06-16 | 2020-09-22 | 桂林电子科技大学 | Gem image fusion method based on image fusion system |
| CN111696067B (en) * | 2020-06-16 | 2023-04-07 | 桂林电子科技大学 | Gem image fusion method based on image fusion system |
| CN111915633A (en) * | 2020-07-24 | 2020-11-10 | 安徽工程大学 | Casting edge detection method and device |
| CN115984316A (en) * | 2023-03-20 | 2023-04-18 | 东莞市兆丰精密仪器有限公司 | Industrial image edge extraction method and device in complex environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107133962A (en) | A kind of diamond saw blade extracting thermal crack method based on rim detection | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN111062915B (en) | Real-time steel pipe defect detection method based on improved YOLOv3 model | |
| CN108074231B (en) | Magnetic sheet surface defect detection method based on convolutional neural network | |
| CN111080622B (en) | Neural network training method, workpiece surface defect classification and detection method and device | |
| CN111879241B (en) | Mobile phone battery size measuring method based on machine vision | |
| CN102175700B (en) | Method for detecting welding seam segmentation and defects of digital X-ray images | |
| CN111402226A (en) | Surface defect detection method based on cascade convolution neural network | |
| CN110473173A (en) | A kind of defect inspection method based on deep learning semantic segmentation | |
| WO2023168972A1 (en) | Linear array camera-based copper surface defect detection method and apparatus | |
| CN107607554A (en) | A kind of Defect Detection and sorting technique of the zinc-plated stamping parts based on full convolutional neural networks | |
| CN104050446B (en) | Meter pointer image identification method based on pointer width character | |
| CN108596880A (en) | Weld defect feature extraction based on image procossing and welding quality analysis method | |
| CN105678767B (en) | A kind of cloth surface flaw detection method based on SoC Hardware/Software Collaborative Design | |
| CN102974551A (en) | Machine vision-based method for detecting and sorting polycrystalline silicon solar energy | |
| CN109284718B (en) | Inspection robot-oriented variable-view-angle multi-instrument simultaneous identification method | |
| CN109540925B (en) | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator | |
| CN110490826B (en) | Fog drop sample image processing method in camera | |
| CN106557751A (en) | A kind of pointer pressure automatic reading method with color configured information | |
| CN107490583A (en) | A kind of intermediate plate defect inspection method based on machine vision | |
| CN112819844B (en) | Image edge detection method and device | |
| CN107169967A (en) | A kind of diamond saw blade crack detecting method based on image procossing | |
| CN106018179A (en) | Glue material surface viscosity measurement method and device based on image processing | |
| CN104952754A (en) | Coated silicon chip sorting method based on machine vision | |
| CN112489042A (en) | Metal product printing defect and surface damage detection method based on super-resolution reconstruction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170905 |