CN102305798A - Method for detecting and classifying glass defects based on machine vision - Google Patents
Method for detecting and classifying glass defects based on machine vision Download PDFInfo
- Publication number
- CN102305798A CN102305798A CN201110219599A CN201110219599A CN102305798A CN 102305798 A CN102305798 A CN 102305798A CN 201110219599 A CN201110219599 A CN 201110219599A CN 201110219599 A CN201110219599 A CN 201110219599A CN 102305798 A CN102305798 A CN 102305798A
- Authority
- CN
- China
- Prior art keywords
- value
- target area
- pixel
- detection
- glass
- 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
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a method for detecting and classifying glass defects based on machine vision. The method comprises the following steps of: extracting a defect area in a picture provided by a camera (linear scanning) by Canny edge detection to obtain a minimum connected domain of the defects; processing a target area according to a filter and a W characteristic which are provided by the method; defining nine kinds of characteristic modes, scanning the minimum connected domain according to rows and columns, and counting the emerging frequency of the nine kinds of characteristic modes in a sample; and judging the types of the defects (bubbles are used as hollow defects, and impurities are used as solid defects) on the basis of the emerging frequency. Compared with the prior art, the machine-vision-based method has the advantages of simple algorithm, quick operation speed, high accuracy and the like, and a new reliable method is provided for the detection of the glass defects.
Description
Technical field
The present invention relates to glass defect and detect sorting technique, relate to a kind of detection and sorting technique of the glass defect based on machine vision particularly.
Background technology
In the commercial production, because various technology or production technology problem can cause certain defective.In the glass production process, just have bubble or impurity and introduce.Different defects is different to the influence of product quality.Possibly influence not quite for common domestic glass such as air blister defect, and very big for the performance impact of the safety glass of automobile.From on May 1st, 2003, China carried out forced examination to automotive safety glass, building safety glass, rail truck with safety glass.And for the very big product of surface area, only depending on manual work to remove defect recognition obviously is not a kind of high-efficiency method.For fear of the loss that manual detection erroneous judgement causes, effectively reduce production costs and promote the percentage of A-class goods, detection and classification problem that the appliance computer vision solves defective become a kind of trend gradually.Because the precision and the current conditions problem of camera, the resolution of the defect image that sometimes obtains is very low.Some defective has only the 4-6 pixel size.On the image of this size, the effective characteristic of difficult acquisition.Therefore classifying through the gray value information of desired value is a way.
Through the discovery of searching to the prior art document, some researchers' work at present concentrates on the number of searching of defective and statistical shortcomings.People such as Zhou Xueqin were published in the paper on " microcomputer information " " based on the detection of the glass blister of regional area threshold value " and have adopted following method in 2007: at first ask variance image and binaryzation to confirm the regional area that bubble belongs in image; In this regional area; Can think that illumination is uniform; Use the regional area threshold method to keep brighter target then, thereby extract bubble one by one.Be published in 2008 in An online defects inspection method for float glass fabrication based on machine vision on the International Journal of Advanced Manufacturing Technology (advanced manufacturing technology international periodical) (a kind of defect online detection method of the float glass manufacturing process based on the machine vision) paper people such as Peng and adopted the OTSU method to come realization prospect and background segment.Seldom a part of researcher has the type of further research defective.People such as Han were published in the circularity that information such as the area, border of the defective that utilization extracts among the paper A study on enhanced algorithms for detecting defects of glasses (the Enhancement Method research of glass defect monitoring) of International Conference on Convergence and Hybrid Information Technology (convergence with hybrid technology international conference) is studied defective in 2009.People such as Hu be published in 2009 InternationalConferenceonComputationalIntelligenceandNat uralComputing(computer intelligence with naturally calculating international conference) paper Analgorithmofglassimagerecognitionbasedonwaveletpacketde composition(glass image recognition algorithm based on WAVELET PACKET DECOMPOSITION) in adopted WAVELET PACKET DECOMPOSITION to realize binaryzation to defect image; Utilize the area of defect area again; Length-width ratio; Gray average and variance, methods such as circularity attempt classifying to defective.They experimentize the rate of accuracy reached to 93.3% of last test sample book to 210 samples (wherein 150 are used for study).Though this method can realize the detection of defective and classification, but the very few truth that cannot embody industrial production environment fully of training sample, and algorithm speed neither be very satisfactory in addition.
Summary of the invention
The present invention is specifically related to a kind of from glass extracting target from images zone and carry out Flame Image Process, then according to handling the method that characteristic information that the back obtains is classified defect type.Can be applicable to industrial goods surface defects detection and identification, belong to the classification problem in the pattern-recognition.The objective of the invention is to glass surface defects detection problem, propose a kind of quick determination methods of the defect type based on Flame Image Process.This method can be even at uneven illumination, and other light sources disturbs the low resolution defect image that obtains under the backgrounds such as (like the lens that form in the glass blister) to carry out signature analysis, thereby judge the type of defective.
Be to realize that above-mentioned purpose, the present invention at first extract the defect area in the picture that camera (line sweep) provides, thereby obtain the minimum connected domain of target.Afterwards binary conversion treatment is carried out in the target area.By the minimum connected domain of column scan, add up the number of 9 types of two-value feature modes (hereinafter will introduce) by row.Judge the type (hollow is bubble, and solid is impurity) of defective on this basis.
According to an aspect of the present invention, a kind of detection and sorting technique of the glass defect based on machine vision is provided, it is characterized in that, comprise the steps:
Step 1: image is carried out the defective rim detection to obtain the marginal information of defective, confirm the target area according to said marginal information;
Step 2: binary conversion treatment is carried out in said target area;
Step 3: remove the noise spot in the said target area;
Step 4: the number of times according to certain row gray-scale value saltus step defines 9 category feature patterns;
Step 5: extract two-value characteristic sequence histogram; Binary image for the said target area that obtains is sought said 9 category feature patterns line by line, thereby adds up said 9 category feature patterns are accomplished the target defect type in the frequency of said target area appearance judgement.
Preferably; In said step 1; Utilize the glass that will detect picture pick-up device to read in computing machine; Utilize rim detection to obtain the marginal information of defective; With the distance of the point on the left side of each defective and the point on the right side width, with the distance of uppermost point of each defective and nethermost point height as said target area as said target area.
Preferably, in said step two, for each pixel
seeking information about their local gray value
, 476.3025 values, with
record binarized gray value information, including:
Where,
is the (m, n) pixel gray value,
is the right value, where the weights templates are as follows:
Wherein, whether certain pixel exists the determination methods of W characteristic to comprise: ask following four eigenwerts earlier:
Then, if WFV11, WFV12, WFV21, WFV22 satisfy in following two conditions, so just judge that there is the W characteristic in this pixel:
Condition 1: (WFV11>2 & WFV12>2) | (WFV21>2 & WFV22>2)
Condition 2: (WFV11>1 & WFV12>1) & (WFV21>1 & WFV22>1).
Preferably, in said step 4 and step 5, each the row extraction two-value characteristic sequence L to the target area after the binaryzation comprises under the following substep:
Substep 1: make the 1st of sequence L coding L[1] equal the gray-scale value b (i, 1) of this first pixel of row, put k=1, j=2;
Sub Step 2: Read the row j-th pixel gray value
, if b (i, j) and L [k] are different, update the characteristic sequence of binary L [k +1] = b (i, j) , set k: = k +1; otherwise, then set
, to perform the sub-step 2 until the two values of the line feature extraction sequence;
Substep 3:, upgrade the number of times that corresponding two-value feature mode occurs according to the length of the two-value characteristic sequence of this row.
The present invention comprising a little compared with prior art: have that algorithm is simple, a fast operation, degree of accuracy advantages of higher, for the industrial products surface defects detection provides a kind of new reliable method.
Description of drawings
Fig. 1 is the inventive method process flow diagram;
Fig. 2 is the image binaryzation method process flow diagram;
Fig. 3 is a binaryzation synoptic diagram as a result;
Fig. 4 is initial gray level image and two-value feature mode recognition result synoptic diagram.
Embodiment
Below in conjunction with accompanying drawing and embodiment technical scheme of the present invention is done further explain.Following examples provided detailed embodiment and process, but protection scope of the present invention are not limited to following embodiment being to implement under the prerequisite with technical scheme of the present invention.
In one embodiment of the invention, the flow process of said method as shown in Figure 1, present embodiment practical implementation step is (preferably use C Plus Plus programming) as follows:
Step 1: extract target area in the image, particularly, image is carried out the marginal information that the defective rim detection obtains defective, confirm the target area according to said marginal information:
Because the defect area in the image (size is near 25 pixel *, 25 pixels) (regard prospect as, size is near 5 pixel *, 5 pixels) is bigger with background difference, therefore can be easier to obtain defect area.Utilize the Canny rim detection can obtain the marginal information of defective.Further can obtain minimum connected domain (rectangle).The extraction of target area width and height: with the distance of the point on the left side of each defect area and the point on the right side width, with the distance of uppermost point and nethermost point height as this defective as this defective.
Step 2: binary conversion treatment is carried out in said target area:
Idiographic flow as shown in Figure 2.For each pixel
seeking information about their local gray value
, 476.3025 values, with
record binarized gray value information, wherein:
Whether certain pixel exists the determination methods of W characteristic following.Ask following four eigenwerts earlier:
If WFV11, WFV12, WFV21, WFV22 satisfy one in following two conditions, so just judge that there is the W characteristic in this pixel.
Condition 1: (WFV11>2 & WFV12>2) | (WFV21>2 & WFV22>2)
Condition 2: (WFV11>1 & WFV12>1) & (WFV21>1 & WFV22>1)
It is 0 that the present invention composes the pixel value of prospect part in the target area, and it is 1 that the pixel value of remaining area is composed, and obtains the binary image of target area, as shown in Figure 3.
Step 3: remove the noise spot in the said target area:
Because the target area is less, therefore adopt following method to remove noise spot.For each pixel
, if
, calculated
Step 4: extract two-value characteristic sequence histogram:
Each row (row) to the binaryzation zone extracts two-value characteristic sequence L, and it comprises the steps:
The 1st coding of substep 1: sequence L equals the gray-scale value of this first pixel of row, L[1]=b (i, 1)
Put k=1; J=2;
Sub Step 2: Read the row j-th pixel gray value
If b (i, j) and L[k] inconsistent, upgrade two-value characteristic sequence L[k+1]=(i j), puts k:=k+1 to b;
Otherwise, set
Step 2 until the completion of the line to extract the binary sequence of characters.
Substep 3:, upgrade the number of corresponding two-value feature mode according to the length of the two-value characteristic sequence of this row.Two-value characteristic sequence length like certain row is 5, and then the number of [10101] two-value feature mode increases by 1.
After the two-value characteristic sequence extraction of completion to all row of row, the number of these 9 types of two-value feature modes in the statistical objects zone.In the sample of Fig. 3 left side, co-exist in 3 [1] patterns, 7 [10101] patterns, 12 [101] patterns, other number of modes are 0 (proper vector of this sample is (3,0,7,12,0,0,0,0,0)).Gray Level Jump situation like the 3rd quasi-mode is the black or white black and white of black and white.
Step 5: judge defect type according to two-value characteristic sequence histogram: we train the proper vector of 9 dimensions of 1000 samples through machine learning classification algorithm AdaBoost, and obtain following reasonably sorter: the target that has [10101] two-value characteristic sequence more than is an air blister defect.
Experimental result shows that the accuracy of the sorting technique of utilizing the feature histogram that the present invention proposes is 93%.Be 0.5ms (the program run environment is the Visual Studio 2008 of OpenCV2.0 and Microsoft company) average calculating operation time of single sample.
Claims (5)
1. detection and sorting technique based on the glass defect of machine vision is characterized in that, comprise the steps:
Step 1: image is carried out the defective rim detection to obtain the marginal information of defective, confirm the target area according to said marginal information;
Step 2: binary conversion treatment is carried out in said target area;
Step 3: remove the noise spot in the said target area;
Step 4: the number of times according to certain row gray-scale value saltus step defines 9 category feature patterns;
Step 5: extract two-value characteristic sequence histogram; Binary image for the said target area that obtains is sought said 9 category feature patterns line by line, thereby adds up said 9 category feature patterns are accomplished the target defect type in the frequency of said target area appearance judgement.
2. the detection and the sorting technique of the glass defect based on machine vision according to claim 1; It is characterized in that; In said step 1; Utilize the glass that will detect picture pick-up device to read in computing machine; Utilize rim detection to obtain the marginal information of defective; With the distance of the point on the left side of each defective and the point on the right side width, with the distance of uppermost point of each defective and nethermost point height as said target area as said target area.
3 according to claim 1, wherein the glass-based machine vision defect detection and classification method, wherein, in the step II, for each pixel
seeking information about their local gray value
476.3025 values, with
record binarized gray value information, wherein:
Where,
the (m, n) pixel gray values,
is the right value, where the weights templates are as follows:
Wherein, whether certain pixel exists the determination methods of W characteristic to comprise: ask following four eigenwerts earlier:
Then, if WFV11, WFV12, WFV21, WFV22 satisfy in following two conditions, so just judge that there is the W characteristic in this pixel:
Condition 1: (WFV11>2 & WFV12>2) | (WFV21>2 & WFV22>2)
Condition 2: (WFV11>1 & WFV12>1) & (WFV21>1 & WFV22>1).
5. according to the detection and the sorting technique of each described glass defect based on machine vision in the claim 1 to 4, it is characterized in that, in said step 4, each row of the target area after the binaryzation extracted two-value characteristic sequence L, comprise following substep:
Substep 1: make the 1st of sequence L coding L[1] equal the gray-scale value b (i, 1) of this first pixel of row, put k=1, j=2;
Sub-Step 2: Read the line j-th pixel gray value
, if b (i, j) and L [k] are different, update the characteristic sequence of binary L [k +1] = b (i, j), set k: = k +1; otherwise, set
, for the sub-step 2 until the value of the bank's two characteristic sequence of extraction;
Substep 3:, upgrade the number of times that corresponding two-value feature mode occurs according to the length of the two-value characteristic sequence of this row.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110219599 CN102305798B (en) | 2011-08-02 | 2011-08-02 | Method for detecting and classifying glass defects based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110219599 CN102305798B (en) | 2011-08-02 | 2011-08-02 | Method for detecting and classifying glass defects based on machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102305798A true CN102305798A (en) | 2012-01-04 |
CN102305798B CN102305798B (en) | 2013-05-22 |
Family
ID=45379679
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201110219599 Expired - Fee Related CN102305798B (en) | 2011-08-02 | 2011-08-02 | Method for detecting and classifying glass defects based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102305798B (en) |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103163156A (en) * | 2013-03-21 | 2013-06-19 | 万新光学集团有限公司 | Automatic grading method of lens defects based on machine vision technology |
CN103759644A (en) * | 2014-01-23 | 2014-04-30 | 广州市光机电技术研究院 | Separating and refining type intelligent optical filter surface defect detecting method |
CN103942777A (en) * | 2014-03-13 | 2014-07-23 | 华南理工大学 | Mobile phone glass cover plate defect detecting method based on principal component analysis |
CN105894036A (en) * | 2016-04-19 | 2016-08-24 | 武汉大学 | Image feature template matching method being applied to detection of mobile phone screen defects |
CN107111872A (en) * | 2014-12-12 | 2017-08-29 | 佳能株式会社 | Information processor, information processing method and program |
CN107169965A (en) * | 2017-06-13 | 2017-09-15 | 福耀集团(上海)汽车玻璃有限公司 | A kind of method and device of inspection glass |
CN107169957A (en) * | 2017-04-28 | 2017-09-15 | 燕山大学 | A kind of glass flaws on-line detecting system and method based on machine vision |
CN108445020A (en) * | 2018-01-31 | 2018-08-24 | 彩虹显示器件股份有限公司 | A kind of glass substrate defect aggregation recognition methods |
CN108982544A (en) * | 2018-06-20 | 2018-12-11 | 青岛联合创智科技有限公司 | A kind of printed circuit board flaw component detection method |
CN109975308A (en) * | 2019-03-15 | 2019-07-05 | 维库(厦门)信息技术有限公司 | A kind of surface inspecting method based on deep learning |
CN110267893A (en) * | 2017-02-09 | 2019-09-20 | 玻璃技术公司 | The system and correlation technique of optical characteristics for on-line measurement glass plate |
CN110441319A (en) * | 2019-09-09 | 2019-11-12 | 凌云光技术集团有限责任公司 | A kind of detection method and device of open defect |
CN110596120A (en) * | 2019-09-06 | 2019-12-20 | 深圳新视智科技术有限公司 | Glass boundary defect detection method, device, terminal and storage medium |
CN111047574A (en) * | 2019-12-12 | 2020-04-21 | 韦士肯(厦门)智能科技有限公司 | Visual inspection defect identification method for cylindrical workpiece |
US10976152B2 (en) | 2017-01-05 | 2021-04-13 | National Taiwan Normal University | Method for defect inspection of transparent substrate by integrating interference and wavefront recording to reconstruct defect complex images information |
US10983478B2 (en) | 2018-03-22 | 2021-04-20 | National Taiwan Normal University | Complex defect diffraction model and method for defect inspection of transparent substrate |
CN113327257A (en) * | 2021-06-02 | 2021-08-31 | 蚌埠凯盛工程技术有限公司 | Method and device for judging automobile glass in different areas |
CN114612385A (en) * | 2022-01-31 | 2022-06-10 | 南通市通州区锦都拉链有限公司 | Zipper selvage defect identification method based on template matching |
CN115439663A (en) * | 2022-11-07 | 2022-12-06 | 山东经发孔府宴酒业有限公司 | Image data-based lotus leaf Daqu quality identification method |
CN116993715A (en) * | 2023-09-18 | 2023-11-03 | 山东庆葆堂生物科技有限公司 | Glass bottle air tightness detection method |
CN117455778A (en) * | 2023-12-21 | 2024-01-26 | 日照市茂源电子有限责任公司 | Optical glass prefabricated member detection method and system based on image enhancement |
CN117491391A (en) * | 2023-12-29 | 2024-02-02 | 登景(天津)科技有限公司 | Glass substrate light three-dimensional health detection method and equipment based on chip calculation |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1687761A (en) * | 2005-04-21 | 2005-10-26 | 上海交通大学 | Automatic method for recognizing pulverization of zinc coated based on machine vision |
JP2006220528A (en) * | 2005-02-10 | 2006-08-24 | Toppan Printing Co Ltd | System and method for defect classification |
US20100150426A1 (en) * | 2004-09-29 | 2010-06-17 | Dainippon Screen Mfg. Co., Ltd. | Apparatus and method for inspecting pattern |
CN101996405A (en) * | 2010-08-30 | 2011-03-30 | 中国科学院计算技术研究所 | Method and device for rapidly detecting and classifying defects of glass image |
-
2011
- 2011-08-02 CN CN 201110219599 patent/CN102305798B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100150426A1 (en) * | 2004-09-29 | 2010-06-17 | Dainippon Screen Mfg. Co., Ltd. | Apparatus and method for inspecting pattern |
JP2006220528A (en) * | 2005-02-10 | 2006-08-24 | Toppan Printing Co Ltd | System and method for defect classification |
CN1687761A (en) * | 2005-04-21 | 2005-10-26 | 上海交通大学 | Automatic method for recognizing pulverization of zinc coated based on machine vision |
CN101996405A (en) * | 2010-08-30 | 2011-03-30 | 中国科学院计算技术研究所 | Method and device for rapidly detecting and classifying defects of glass image |
Non-Patent Citations (3)
Title |
---|
DEREK C. SCHUURMAN ET AL.: "Real-Time Synchronized Vision Sensors Over Ethernet", 《LOCAL COMPUTER NETWORKS》, 18 November 2004 (2004-11-18) * |
余文勇 等: "一种浮法玻璃全面缺陷在线检测系统", 《华中科技大学学报(自然科学版)》, vol. 35, no. 8, 31 August 2007 (2007-08-31) * |
王飞 等: "一种平板玻璃缺陷在线检测系统的研究", 《应用光学》, vol. 31, no. 1, 31 January 2010 (2010-01-31) * |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103163156A (en) * | 2013-03-21 | 2013-06-19 | 万新光学集团有限公司 | Automatic grading method of lens defects based on machine vision technology |
CN103759644A (en) * | 2014-01-23 | 2014-04-30 | 广州市光机电技术研究院 | Separating and refining type intelligent optical filter surface defect detecting method |
CN103759644B (en) * | 2014-01-23 | 2016-05-18 | 广州市光机电技术研究院 | A kind of separation refinement intelligent detecting method of optical filter blemish |
CN103942777A (en) * | 2014-03-13 | 2014-07-23 | 华南理工大学 | Mobile phone glass cover plate defect detecting method based on principal component analysis |
CN107111872B (en) * | 2014-12-12 | 2021-01-12 | 佳能株式会社 | Information processing apparatus, information processing method, and storage medium |
CN107111872A (en) * | 2014-12-12 | 2017-08-29 | 佳能株式会社 | Information processor, information processing method and program |
CN105894036A (en) * | 2016-04-19 | 2016-08-24 | 武汉大学 | Image feature template matching method being applied to detection of mobile phone screen defects |
CN105894036B (en) * | 2016-04-19 | 2019-04-09 | 武汉大学 | A kind of characteristics of image template matching method applied to mobile phone screen defects detection |
US10976152B2 (en) | 2017-01-05 | 2021-04-13 | National Taiwan Normal University | Method for defect inspection of transparent substrate by integrating interference and wavefront recording to reconstruct defect complex images information |
CN110267893B (en) * | 2017-02-09 | 2022-05-13 | 玻璃技术公司 | System for on-line measurement of optical properties of glass sheets and related method |
CN110267893A (en) * | 2017-02-09 | 2019-09-20 | 玻璃技术公司 | The system and correlation technique of optical characteristics for on-line measurement glass plate |
CN107169957A (en) * | 2017-04-28 | 2017-09-15 | 燕山大学 | A kind of glass flaws on-line detecting system and method based on machine vision |
CN107169965A (en) * | 2017-06-13 | 2017-09-15 | 福耀集团(上海)汽车玻璃有限公司 | A kind of method and device of inspection glass |
CN108445020A (en) * | 2018-01-31 | 2018-08-24 | 彩虹显示器件股份有限公司 | A kind of glass substrate defect aggregation recognition methods |
CN108445020B (en) * | 2018-01-31 | 2021-11-09 | 彩虹显示器件股份有限公司 | Glass substrate defect aggregation identification method |
US10983478B2 (en) | 2018-03-22 | 2021-04-20 | National Taiwan Normal University | Complex defect diffraction model and method for defect inspection of transparent substrate |
CN108982544A (en) * | 2018-06-20 | 2018-12-11 | 青岛联合创智科技有限公司 | A kind of printed circuit board flaw component detection method |
CN109975308B (en) * | 2019-03-15 | 2023-03-24 | 维库(厦门)信息技术有限公司 | Surface detection method based on deep learning |
CN109975308A (en) * | 2019-03-15 | 2019-07-05 | 维库(厦门)信息技术有限公司 | A kind of surface inspecting method based on deep learning |
CN110596120A (en) * | 2019-09-06 | 2019-12-20 | 深圳新视智科技术有限公司 | Glass boundary defect detection method, device, terminal and storage medium |
CN110441319A (en) * | 2019-09-09 | 2019-11-12 | 凌云光技术集团有限责任公司 | A kind of detection method and device of open defect |
CN111047574A (en) * | 2019-12-12 | 2020-04-21 | 韦士肯(厦门)智能科技有限公司 | Visual inspection defect identification method for cylindrical workpiece |
CN113327257A (en) * | 2021-06-02 | 2021-08-31 | 蚌埠凯盛工程技术有限公司 | Method and device for judging automobile glass in different areas |
CN113327257B (en) * | 2021-06-02 | 2022-07-26 | 蚌埠凯盛工程技术有限公司 | Method and device for judging automobile glass in different areas |
CN114612385A (en) * | 2022-01-31 | 2022-06-10 | 南通市通州区锦都拉链有限公司 | Zipper selvage defect identification method based on template matching |
CN115439663A (en) * | 2022-11-07 | 2022-12-06 | 山东经发孔府宴酒业有限公司 | Image data-based lotus leaf Daqu quality identification method |
CN115439663B (en) * | 2022-11-07 | 2023-02-17 | 山东经发孔府宴酒业有限公司 | Image data-based lotus leaf Daqu quality identification method |
CN116993715A (en) * | 2023-09-18 | 2023-11-03 | 山东庆葆堂生物科技有限公司 | Glass bottle air tightness detection method |
CN116993715B (en) * | 2023-09-18 | 2023-12-19 | 山东庆葆堂生物科技有限公司 | Glass bottle air tightness detection method |
CN117455778A (en) * | 2023-12-21 | 2024-01-26 | 日照市茂源电子有限责任公司 | Optical glass prefabricated member detection method and system based on image enhancement |
CN117455778B (en) * | 2023-12-21 | 2024-04-05 | 日照市茂源电子有限责任公司 | Optical glass prefabricated member detection method and system based on image enhancement |
CN117491391A (en) * | 2023-12-29 | 2024-02-02 | 登景(天津)科技有限公司 | Glass substrate light three-dimensional health detection method and equipment based on chip calculation |
CN117491391B (en) * | 2023-12-29 | 2024-03-15 | 登景(天津)科技有限公司 | Glass substrate light three-dimensional health detection method and equipment based on chip calculation |
Also Published As
Publication number | Publication date |
---|---|
CN102305798B (en) | 2013-05-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102305798B (en) | Method for detecting and classifying glass defects based on machine vision | |
CN108982508B (en) | Plastic package IC chip defect detection method based on feature template matching and deep learning | |
CN101996405B (en) | Method and device for rapidly detecting and classifying defects of glass image | |
CN107437243B (en) | Tire impurity detection method and device based on X-ray image | |
CN106570511B (en) | Keyboard character defect online detection system and method based on computer vision | |
CN109978822B (en) | Banana maturity judging modeling method and judging method based on machine vision | |
CN104992449B (en) | Information identification and surface defect online test method based on machine vision | |
CN104574353B (en) | The surface defect decision method of view-based access control model conspicuousness | |
CN109035195B (en) | Fabric defect detection method | |
CN109308700A (en) | A kind of visual identity defect inspection method based on printed matter character | |
CN105447503B (en) | Pedestrian detection method based on rarefaction representation LBP and HOG fusion | |
CN106934386B (en) | A kind of natural scene character detecting method and system based on from heuristic strategies | |
CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
CN103903018A (en) | Method and system for positioning license plate in complex scene | |
CN114972356B (en) | Plastic product surface defect detection and identification method and system | |
CN104050684B (en) | A kind of video frequency motion target sorting technique based on on-line training and system | |
CN109540925B (en) | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator | |
CN105388162A (en) | Raw material silicon wafer surface scratch detection method based on machine vision | |
Liu et al. | A computer vision system for automatic steel surface inspection | |
CN103310211A (en) | Filling mark recognition method based on image processing | |
CN104198497A (en) | Surface defect detection method based on visual saliency map and support vector machine | |
CN113177924A (en) | Industrial production line product flaw detection method | |
CN104952754B (en) | Silicon chip method for separating after plated film based on machine vision | |
CN106780437B (en) | A kind of quick QFN chip plastic packaging image obtains and amplification method | |
CN103759644A (en) | Separating and refining type intelligent optical filter surface defect detecting method |
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: 20130522 Termination date: 20150802 |
|
EXPY | Termination of patent right or utility model |