CN103175839A - Processing method and system for detection of offset plate surface - Google Patents
Processing method and system for detection of offset plate surface Download PDFInfo
- Publication number
- CN103175839A CN103175839A CN2011104329940A CN201110432994A CN103175839A CN 103175839 A CN103175839 A CN 103175839A CN 2011104329940 A CN2011104329940 A CN 2011104329940A CN 201110432994 A CN201110432994 A CN 201110432994A CN 103175839 A CN103175839 A CN 103175839A
- Authority
- CN
- China
- Prior art keywords
- flaw
- image
- offset plate
- plate material
- original image
- 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
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a processing method and a processing system for detection of an offset plate surface, and relates to the field of image processing technologies. The method comprises the following steps of: S1, carrying out boundary location on the original image of the offset plate surface in order to gain an effective area in the original image, and using the effective area as an image to be detected; S2, preprocessing the image to be detected; S3, distinguishing the defects on the preprocessed image to be detected in order to gain the positions of the defects; and S4, gaining the characteristic parameters of the defects according to the positions of the defects. According to the processing method and the processing system for the detection of the offset plate surface, disclosed by the invention, the machine vision technology based on the image processing technology is adopted, the detection is carried out based on the machine vision, and therefore, the detection speed and the detection precision for the defects on the surface of an offset plate can be improved, and the cost is decreased.
Description
Technical field
The present invention relates to technical field of image processing, particularly a kind of disposal route and system of offset plate material surface detection.
Background technology
Offset plate material (offset printing plate), process through coating on the surface, can produce the zone of transfer of ink and the area surface of non-transfer of ink thereon, and two kinds of zones is in conplane flat panel workpieces.At present, the general trend in China's offset plate material market is " PS version output rapidly improves, and CTP plate new varieties continue to bring out; market competition is fiercer ", but produce and use along with offset plate material is large-scale, for a long time, the quality problems of offset plate material are also come many.
At present, in the production line, be mainly to rely on experience to come directly the quality of offset plate material to be observed by the workman, be to detect in conjunction with the quality of the printed matter quality to offset plate material.This detection mode speed is low, precision is low and cost is high.
Summary of the invention
The technical matters that (one) will solve
The technical problem to be solved in the present invention is: how to improve detection speed and the accuracy of detection of the flaw on offset plate material surface, and reduce costs.
(2) technical scheme
For solving the problems of the technologies described above, the invention provides the disposal route of a kind of offset plate material surface detection, said method comprising the steps of:
S1: the original image to the offset plate material surface carries out boundary alignment, obtaining the effective coverage in described original image, and with described effective coverage as testing image;
S2: described testing image is carried out pre-service;
S3: pretreated testing image is carried out flaw identification, to obtain the position of flaw;
S4: according to the position of flaw, obtain the characterisitic parameter of flaw.
Preferably, in step S1, the original image on offset plate material surface obtains by a linear array CCD camera collection.
Preferably, the original image that in step S1, offset plate material shows obtains by the linear array CCD camera collection of two parallel splicings at least;
Also comprise step between step S1 and step S2:
S21: the testing image that the linear array CCD cameras of at least two parallel splicings are obtained respectively, and the intersection between every two testing images that will obtain only is attributed to a testing image and processes.
Preferably, in step S2, described pre-service is adopted: at least a in figure image intensifying, Gauss's denoising, dual threshold algorithm and morphology.
Preferably, the dual threshold algorithm that adopts in step S2 is the dual threshold algorithm of adaptive threshold.
Preferably, when carrying out pre-service in step S2, pretreated testing image is divided at least two surveyed areas, each surveyed area is adopted the parameter of different dual threshold algorithms.
Preferably, described in step S4, the characterisitic parameter of flaw comprises: at least one in the area of described flaw, length and width, gray average, boundary rectangle length and width, flaw coordinate and flaw type.
The invention also discloses the disposal system of a kind of offset plate material surface detection, described system comprises:
The boundary alignment module is used for the original image on offset plate material surface is carried out boundary alignment, obtaining the effective coverage in described original image, and with described effective coverage as testing image;
Pretreatment module is used for described testing image is carried out pre-service;
The flaw identification module is used for pretreated testing image is carried out flaw identification, to obtain the position of flaw;
The gain of parameter module is used for the position according to flaw, obtains the characterisitic parameter of flaw.
(3) beneficial effect
The present invention detects by machine vision by take the machine vision technique of image processing techniques as the basis, has improved detection speed and the accuracy of detection of the flaw on offset plate material surface, and has reduced cost.
Description of drawings
Fig. 1 is the image collecting device structural representation that is used for image acquisition according to the offset plate material surface detection of one embodiment of the present invention;
Fig. 2 is the process flow diagram according to the disposal route of the offset plate material surface detection of one embodiment of the present invention;
Fig. 3 is the splicing schematic diagram of the image that gathers respectively of four cameras.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used for explanation the present invention, but are not used for limiting the scope of the invention.
In present embodiment, with reference to Fig. 1, described system illustrates the collection of image take a camera as example, and described image collecting device comprises: camera 1, image processing equipment 6 and light source 3, and described camera 1 gathers the original image of PS version 5 upper surfaces; Described image processing equipment is processed the original image that described camera gathers, to identify the flaw 2 of described PS version upper surface; Described light source 3 is provided as the picture illumination when gathering the original image of PS version 5 upper surfaces for described camera 1.
In present embodiment, by linear array charge coupled cell (Charge-coupled Device, CCD) collected by camera offset plate material image, in 30 pixels of the many collections of offset plate material left and right edges meeting, guarantee to gather the integrality of image, the pixels that gather (or be detection platform or be ground) need to be carried out boundary alignment, in order to search the effective coverage of offset plate material more at the beginning of detection.
Fig. 2 is the process flow diagram according to the disposal route of the offset plate material surface detection of one embodiment of the present invention; With reference to Fig. 2, the method for present embodiment comprises the following steps:
S1: the original image to the offset plate material surface carries out boundary alignment, obtaining the effective coverage in described original image, and with described effective coverage as testing image;
because the effective coverage of offset plate material image accounts for large many parts, effective coverage and borderline region have notable difference on color range, determine boundary threshold, utilize the boundary search algorithm can obtain boundary coordinate (x, y), algorithm is done following explanation: boundary threshold is made as M, count and be made as N (initial value n=0) in the border that expectation is searched, the long L of image, wide is W, when offset plate material transmits under rolling the effect of taking out, the basic maintenance with direction of transfer in the same way without tilting, during the image calculation left margin, can think temporarily that N the frontier point Y coordinate of choosing is consistent, the Analysis of X coordinate gets final product.In the ideal case, detection platform or ground do not have interfering picture border noiseless, the coordinate of N frontier point is done simple mean value computation just can obtain boundary coordinate, can noise totally not occur due to airborne dust, ground, and simple average affects whole detection.So, try not to use the method in order to guarantee accuracy.For this reason, the X coordinate of N frontier point is done a bubble sort, in the middle of then choosing, 3 data are done average, it is poor that point range after then sorting and average are done, reading in scope, scope other places rejecting, so just obtained left margin X coordinate comparatively accurately, straight-line equation is Y=X.
S2: described testing image is carried out pre-service;
S3: pretreated testing image is carried out flaw identification, to obtain the position of flaw;
S4: according to the position of flaw, obtain the characterisitic parameter of flaw, in present embodiment, the characterisitic parameter of described flaw comprises: at least one in the area of described flaw, length and gray average; First the position according to flaw positions flaw, then according to features such as the area of each edge calculations flaw, length and width, gray averages, area calculates the area that normally calculates the flaw outline polygon, if but flaw is not to fill full whole profile, calculate just inaccurate.Therefore designed a kind of new area computation method, first calculate left summit rectX coordinate, the rectY coordinate of the long rectL of the boundary rectangle of flaw profile, wide rectW and boundary rectangle, again with process after image corresponding external, if read pixel for black cumulative, at last to black number of pixels be the flaw area.
In step S1, the original image on offset plate material surface can obtain by a linear array CCD camera collection;
but in order to guarantee accuracy of detection, detect the offset plate material of wide format many line array CCDs need to be installed, there is the image-region with other picture registration on the left margin of each image or right margin or border, left and right like this, this just needs to determine to overlap the zone, image after guaranteeing is processed not double counting, flaw quantity does not repeat to add up, directly intersection only being classified as an image goes to calculate, another is clipped corresponding image information and gets final product, preferably, the original image that in step S1, offset plate material shows obtains by the linear array CCD camera collection of two parallel splicings at least, accordingly, also comprise step between step S1 and step S2:
S21: the testing image that the linear array CCD cameras of at least two parallel splicings are obtained respectively, and the intersection between every two testing images that will obtain only is attributed to a testing image and processes;
If detected fabric width is L, the overlapping pixels number of camera is a, and Pixel size is b, if adopt the 4K camera, required camera number n=L/ (4096*b), if integer or decimal are considered overlapping pixels, advance one; Namely, 4096*n-(n-1) * a>L, there is the image-region with other picture registration on the left margin of each image or right margin or border, left and right like this, and this just needs to determine to overlap the zone, image after guaranteeing is processed not double counting, and flaw quantity does not repeat to add up.
All data units are pixel.With reference to Fig. 3, suppose that the collection width of every camera is 4000, the edge leaves and takes 30, overlaps 100, and gathering altogether picture traverse is 4000*4-100*3-30*2=15640.Represent a camera, totally 4 every 4000.Directly intersection only is classified as an image and goes to calculate, another is clipped corresponding image information and gets final product.Leftmost border, namely scale 0 place of upper figure, be respectively 30 pixels apart from the offset plate material left margin; Wherein gray area is the left and right intersection.Fabric width refers to the width of every required processing image of camera, border such as the following table that intercept when calculating:
Camera numbers | Gather the border | The details in a play not acted out on stage, but told through dialogues fabric width |
I | (30,4000) | 3970 |
II | (4000,7900) | 3900 |
III | (7900,11800) | 3900 |
IV | (11800,15670) | 3870 |
Add up to | 15640 |
Need described testing image is first carried out pre-service after testing image obtaining, preferably, in present embodiment, in step S2, described pre-service is adopted: at least a in figure image intensifying, Gauss's denoising, dual threshold algorithm and morphology.
Because the flaw on offset plate material surface has the types such as offset printing, dirty point, scuffing, pit, after the linear array CCD scanning image, offset plate material background color range is basically identical when the even no-reflection of illumination, the color range of flaw have than background large also have less than background, this just need to use the dual threshold algorithm, and described dual threshold algorithmic formula is as follows:
Wherein, data[i] [j] be the grey level of the pixel of the capable j of image i row (the gray scale span is 0~255, and gets 0 and be black, gets 255 and is white), the value of max_value is decided to be 255, m and M are respectively the gray threshold of pixel, and the value of m and M depends on the grey level of plate, for example, the plate gray scale is 40 ± 15, think background in (25,55), other are flaw.
If image background is inhomogeneous, again because offset plate material is reflective, cause the grey level of background in a larger scope, the gray-scale value of some flaws also in this scope, is understood lost part flaw information after application dual threshold algorithm.At this moment, the dual threshold algorithm by adaptive threshold can solve problems.Adaptive threshold, it is the local auto-adaptive binaryzation that image is carried out, entire image is divided into several windows, then each window is set the threshold value of this window according to certain rule, so just obtained the image of binaryzation, preferably, the dual threshold algorithm that adopts in step S2 is the dual threshold algorithm of adaptive threshold, the dual threshold algorithm of application self-adapting threshold value had both guaranteed that flaw information can not lose, and the unevenness that has solved again background image is disturbed.
In service at production line, if the same flaw appears in each piece offset plate material in the fixed position, run into this kind situation, the application partition detection algorithm can solve.Subarea detecting is that the offset plate material image is divided into several surveyed areas, at each surveyed area, accuracy of detection and detection algorithm parameter are set, guarantee to detect the flaw of dissimilar, different stage, different gray-scale values, preferably, when running into above-mentioned situation, when carrying out pre-service in step S2, pretreated testing image is divided at least two surveyed areas, each surveyed area is adopted the parameter of different dual threshold algorithms.
The invention also discloses the disposal system of a kind of offset plate material surface detection, described system comprises:
The boundary alignment module is used for the original image on offset plate material surface is carried out boundary alignment, obtaining the effective coverage in described original image, and with described effective coverage as testing image;
Pretreatment module is used for described testing image is carried out pre-service;
The flaw identification module is used for pretreated testing image is carried out flaw identification, to obtain the position of flaw;
The gain of parameter module is used for the position according to flaw, obtains the characterisitic parameter of flaw.
In present embodiment, by take image processing techniques as the basis machine vision technique, detect by machine vision, the automaticity that not only can greatly enhance productivity and produce, and machine vision is easy to realize that information is integrated, satisfy the requirement of digitizing, automated production, the speed that also reaches is high, precision is high and the purpose of saving cost of labor.
Above embodiment only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (8)
1. the disposal route of an offset plate material surface detection, is characterized in that, said method comprising the steps of:
S1: the original image to the offset plate material surface carries out boundary alignment, obtaining the effective coverage in described original image, and with described effective coverage as testing image;
S2: described testing image is carried out pre-service;
S3: pretreated testing image is carried out flaw identification, to obtain the position of flaw;
S4: according to the position of flaw, obtain the characterisitic parameter of flaw.
2. the method for claim 1, is characterized in that, in step S1, the original image on offset plate material surface obtains by a linear array CCD camera collection.
3. the method for claim 1, is characterized in that, the original image that in step S1, offset plate material shows obtains by the linear array CCD camera collection of two parallel splicings at least;
Also comprise step between step S1 and step S2:
S21: the testing image that the linear array CCD cameras of at least two parallel splicings are obtained respectively, and the intersection between every two testing images that will obtain only is attributed to a testing image and processes.
4. the method for claim 1, is characterized in that, in step S2, described pre-service is adopted: at least a in figure image intensifying, Gauss's denoising, dual threshold algorithm and morphology.
5. method as claimed in claim 4, is characterized in that, the dual threshold algorithm that adopts in step S2 is the dual threshold algorithm of adaptive threshold.
6. method as claimed in claim 5, is characterized in that, when carrying out pre-service in step S2, pretreated testing image is divided at least two surveyed areas, each surveyed area adopted the parameter of different dual threshold algorithms.
7. the method for claim 1, is characterized in that, the characterisitic parameter of flaw described in step S4 comprises: at least one in the area of described flaw, length and width, gray average, boundary rectangle length and width, flaw coordinate and flaw type.
8. the disposal system of an offset plate material surface detection, is characterized in that, described system comprises:
The boundary alignment module is used for the original image on offset plate material surface is carried out boundary alignment, obtaining the effective coverage in described original image, and with described effective coverage as testing image;
Pretreatment module is used for described testing image is carried out pre-service;
The flaw identification module is used for pretreated testing image is carried out flaw identification, to obtain the position of flaw;
The gain of parameter module is used for the position according to flaw, obtains the characterisitic parameter of flaw.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011104329940A CN103175839A (en) | 2011-12-21 | 2011-12-21 | Processing method and system for detection of offset plate surface |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2011104329940A CN103175839A (en) | 2011-12-21 | 2011-12-21 | Processing method and system for detection of offset plate surface |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103175839A true CN103175839A (en) | 2013-06-26 |
Family
ID=48635841
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2011104329940A Pending CN103175839A (en) | 2011-12-21 | 2011-12-21 | Processing method and system for detection of offset plate surface |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103175839A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103454283A (en) * | 2013-09-12 | 2013-12-18 | 电子科技大学 | Line detection system and method |
CN105388162A (en) * | 2015-10-28 | 2016-03-09 | 镇江苏仪德科技有限公司 | Raw material silicon wafer surface scratch detection method based on machine vision |
CN109738461A (en) * | 2019-01-04 | 2019-05-10 | 安徽地势坤光电科技有限公司 | A kind of halftone visible detection method applied in screen printing apparatus |
CN110857920A (en) * | 2018-08-24 | 2020-03-03 | 东华大学 | Method for detecting poor forming defect of coiled filament |
CN114742788A (en) * | 2022-04-01 | 2022-07-12 | 南通高精数科机械有限公司 | Copper bar defect detection method and system based on machine vision |
CN117309903A (en) * | 2023-10-10 | 2023-12-29 | 青岛峻海物联科技有限公司 | Method and device for positioning defects in tunnel |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1400922A1 (en) * | 2002-09-20 | 2004-03-24 | Nitto Denko Corporation | Print inspection method and apparatus |
CN1670513A (en) * | 2004-03-17 | 2005-09-21 | 中国印钞造币总公司 | Apparatus and method for detecting sheet-like material |
WO2006088150A1 (en) * | 2005-02-17 | 2006-08-24 | Central Glass Company, Limited | Glass substrate inspection equipment and inspection method |
-
2011
- 2011-12-21 CN CN2011104329940A patent/CN103175839A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1400922A1 (en) * | 2002-09-20 | 2004-03-24 | Nitto Denko Corporation | Print inspection method and apparatus |
CN1670513A (en) * | 2004-03-17 | 2005-09-21 | 中国印钞造币总公司 | Apparatus and method for detecting sheet-like material |
WO2006088150A1 (en) * | 2005-02-17 | 2006-08-24 | Central Glass Company, Limited | Glass substrate inspection equipment and inspection method |
Non-Patent Citations (5)
Title |
---|
《科技博览大百科》编写组: "《百科大揭密 科技博览大百科》", 31 May 2011, article "百科大揭秘 科技博览大百科", pages: 157 * |
余洪山等: "一种改进型Canny边缘检测算法", 《计算机工程与应用》, vol. 40, no. 20, 31 December 2004 (2004-12-31), pages 27 - 29 * |
孙春凤等: "《一种新的边缘保持局部自适应图像插值算法》", 《仪器仪表学报》, vol. 31, no. 10, 31 October 2010 (2010-10-31), pages 2279 - 2284 * |
李钰等: "自适应双阈值Canny算子的图像边缘检测", 《长春工程学院学报(自然科学版)》, vol. 8, no. 3, 31 December 2007 (2007-12-31), pages 44 - 53 * |
李雅静: "基于图像处理的PS版面缺陷检测的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 9, 15 September 2011 (2011-09-15), pages 5 - 43 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103454283A (en) * | 2013-09-12 | 2013-12-18 | 电子科技大学 | Line detection system and method |
CN105388162A (en) * | 2015-10-28 | 2016-03-09 | 镇江苏仪德科技有限公司 | Raw material silicon wafer surface scratch detection method based on machine vision |
CN105388162B (en) * | 2015-10-28 | 2017-12-01 | 镇江苏仪德科技有限公司 | Raw material silicon chip surface scratch detection method based on machine vision |
CN110857920A (en) * | 2018-08-24 | 2020-03-03 | 东华大学 | Method for detecting poor forming defect of coiled filament |
CN109738461A (en) * | 2019-01-04 | 2019-05-10 | 安徽地势坤光电科技有限公司 | A kind of halftone visible detection method applied in screen printing apparatus |
CN114742788A (en) * | 2022-04-01 | 2022-07-12 | 南通高精数科机械有限公司 | Copper bar defect detection method and system based on machine vision |
CN117309903A (en) * | 2023-10-10 | 2023-12-29 | 青岛峻海物联科技有限公司 | Method and device for positioning defects in tunnel |
CN117309903B (en) * | 2023-10-10 | 2024-05-07 | 青岛峻海物联科技有限公司 | Method and device for positioning defects in tunnel |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103175839A (en) | Processing method and system for detection of offset plate surface | |
CN104616275A (en) | Defect detecting method and defect detecting device | |
CN102043950B (en) | Vehicle outline recognition method based on canny operator and marginal point statistic | |
CN104749184A (en) | Automatic optical detection method and automatic optical detection system | |
CN110555467B (en) | Industrial data classification method based on model migration | |
CN101751572A (en) | Pattern detection method, device, equipment and system | |
CN109255787A (en) | Silk ingot scratch detection system and method based on deep learning and image processing techniques | |
CN101320004A (en) | Bamboo strip defect on-line detection method based on machine vision | |
CN104318543A (en) | Board metering method and device based on image processing method | |
CN102974551A (en) | Machine vision-based method for detecting and sorting polycrystalline silicon solar energy | |
CN105913093A (en) | Template matching method for character recognizing and processing | |
CN104574389A (en) | Battery piece chromatism selection control method based on color machine vision | |
CN106056020A (en) | Bar code detection and recognition method and system based on machine vision system | |
CN115131348B (en) | Method and system for detecting textile surface defects | |
CN102184405A (en) | Image acquisition-analysis method | |
CN111242896A (en) | Color printing label defect detection and quality rating method | |
CN101995412B (en) | Robust glass scratch defect detection method and device thereof | |
CN104978748A (en) | Liquid crystal screen defect detection method based on local pixel values | |
CN102305798A (en) | Method for detecting and classifying glass defects based on machine vision | |
CN105388162A (en) | Raw material silicon wafer surface scratch detection method based on machine vision | |
CN116977358B (en) | Visual auxiliary detection method for corrugated paper production quality | |
CN104483320A (en) | Digitized defect detection device and detection method of industrial denitration catalyst | |
CN107102009A (en) | A kind of method of the cylinder spool quality testing based on machine vision | |
CN110648330B (en) | Defect detection method for camera glass | |
CN109523540A (en) | Yarn break detection method based on Hough transformation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130626 |