WO2007062563A1 - Procede d'examen automatique en ligne pour des detections de defauts de surface d'acier pendant le pretraitement de toles d'acier - Google Patents

Procede d'examen automatique en ligne pour des detections de defauts de surface d'acier pendant le pretraitement de toles d'acier Download PDF

Info

Publication number
WO2007062563A1
WO2007062563A1 PCT/CN2006/002403 CN2006002403W WO2007062563A1 WO 2007062563 A1 WO2007062563 A1 WO 2007062563A1 CN 2006002403 W CN2006002403 W CN 2006002403W WO 2007062563 A1 WO2007062563 A1 WO 2007062563A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
defect
steel sheet
detecting
surface defect
Prior art date
Application number
PCT/CN2006/002403
Other languages
English (en)
Chinese (zh)
Inventor
Rentong Han
Li Chen
Fengguo Zhao
Mudi Xiong
Jinghuan Guo
Original Assignee
Bohai Shipbuilding Industry Co., Ltd.
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Bohai Shipbuilding Industry Co., Ltd. filed Critical Bohai Shipbuilding Industry Co., Ltd.
Publication of WO2007062563A1 publication Critical patent/WO2007062563A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/888—Marking defects
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/89—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles
    • G01N21/8914—Investigating the presence of flaws or contamination in moving material, e.g. running paper or textiles characterised by the material examined
    • G01N2021/8918—Metal
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30136—Metal

Definitions

  • the invention relates to an automatic detection method for surface defects of steel plates, in particular to an online automatic detection method for surface defects in the pretreatment process of steel plates. Background technique
  • the steel plate In the shipbuilding process, a large number of steel plates are required. First, the steel plate needs to be pretreated. After the surface descaling and painting process in the steel plate pretreatment process, the surface defects of the steel plate need to be tested before entering the cutting process. If the over-standard steel plate is to be repaired or replaced, the steel plate that meets the standard can be cut into the next process.
  • the traditional method for detecting surface defects of steel plates is to perform visual inspection after the steel plate has been derusted and sprayed by the pretreatment line. After one side of the steel plate is inspected, the steel plate is turned over to inspect the other side. This inspection method is not only inefficient. It is also unsafe, and the missed inspection is inevitable, so that unqualified steel plates will be used on the hull structure, causing quality problems and rework. Summary of the invention
  • the technical problem to be solved by the invention is to timely discover the surface defects exceeding the standard steel plate on the assembly line, and repair and replace the steel plate, which not only can improve the reliability of the steel plate defect inspection, but also can greatly improve the productivity of the steel plate pretreatment process. Online automatic detection method for surface defects.
  • the present invention provides a method for detecting a surface defect of a steel sheet, the method for detecting a surface defect of the steel sheet comprising the steps of: illuminating a surface of the steel sheet with an illumination device; and imaging the surface of the steel sheet by an image forming apparatus to form The image is processed by the image processor, and the surface of the steel sheet is determined to be defective according to the gray scale information of the image.
  • the present invention provides an online automatic detection method for surface defects in a steel sheet pretreatment process.
  • the method adopts an optical imaging technique, and the imaging color and gray scale of the defect point are different from those of other normal points.
  • Image color and gray feature recognition technology is used to identify and classify defect points, and then input the image into the microcomputer for processing.
  • the present invention provides an online automatic surface defect in a steel sheet pretreatment process.
  • a detection method includes a surface illumination of a steel sheet, an image taken, an image processing, and a defect point identification step.
  • Step 1 Lighting the surface of the steel plate, laying out the surface lighting facilities of the steel plate, using a special light source that does not interfere with the imaging, and forming a light field on both sides of the steel plate.
  • Step 2 The image is taken, and multiple sets of independent imaging measuring devices are arranged corresponding to the light fields on both sides of the steel plate. When the steel plate moves at a certain speed in the horizontal direction, the imaging measuring device simultaneously performs real-time imaging on both sides of the steel plate.
  • Step 3 Image processing and defect point identification.
  • the grayscale image signal collector and the chrominance image signal collector respectively collect the imaged grayscale information and chrominance information, and then by the image processor (for example, on an industrial computer)
  • the pre-programmed software analyzes and processes the image, that is, the image grayscale and chroma feature recognition technology is used to identify and classify the defect points. When the result of the identification and classification is a defect point, the marking device is activated to identify the defect point.
  • the above-mentioned steel plate surface illumination, image capture, image processing and defect point identification are loaded on an industrial computer by a pre-programmed driver and image feature database, driving the light source to illuminate and collect the image, and calculating the image of the defect point image to Query the image feature database, identify the defect size and type, and classify it according to the defect depth level, and then judge whether it is a defect point.
  • the marking device records the surface defect and the X, Y axis coordinate position, and automatically starts the alarm.
  • the marking device ejects the liquid identification, and the defect point characteristic information after the completion of the identification is input into the image feature database for self-learning and expansion.
  • the imaging measuring device continuously performs multiple images and arrays on both sides of the online moving steel plate, which ensures that the horizontally running steel plate is detected without any omission.
  • the above defect points can be classified as follows: The defect depth d is greater than 20% of the plate thickness t, and the area exceeds 2% of the plate surface. This part of the plate needs to be replaced as required; 0.07t ⁇ d ⁇ 0.2t (mm), the specified weld repair After smoothing; d ⁇ 0.07t (mm) and d 3mm, specified smoothing; d ⁇ 0.15mm, the rule does not have to be trimmed.
  • the plate width can reach 4m.
  • FIG. 2 is a block diagram of the control system of the present invention.
  • Figure 3 is a flow chart of the computer software system.
  • Figure 4 is a surface view of a steel plate with scratch defects.
  • Figure 5 is a histogram of Figure 4.
  • Figure 6 is the result of the Sobel gradient operator processing of Figure 4.
  • Figure 7 is a histogram comparison of the gradient map before and after grayscale stretching. detailed description
  • the gantry frame 1 is set on the steel plate pretreatment line, the steel plate 3 moves through the gantry frame 1 along the Y axis at a certain speed; the upper and lower beams of the gantry frame 1 are equipped with lighting facilities 4, and the lighting facility 4 mainly has a special light source for high speed strobe and The lens, the light source forms a parallel beam through the lens, and the parallel beam is projected on both sides of the steel plate 3 to form a light field; the light field on both sides of the steel plate 3 is arranged on the gantry 1 upper and lower beams, two or four sets of independent imaging measuring devices composed of cameras are arranged 2.
  • the imaging measuring device 2 changes the area and depth information of the defect area into CCD plane chromaticity and gray scale information, and the upper surface color
  • the image collector 6 collects the image signals of the upper surface imaging systems 13, 14, 15, and 16 respectively, and the image signals of the lower surface imaging systems 17, 18, 19, and 20 are respectively collected by the lower surface color image collector 7.
  • the software is pre-programmed by the industrial computer 8 to analyze and process the image (as the selection industry can be separately selected by the gray image signal collector)
  • the grayscale imaging information on the upper surface of the steel plate and the grayscale imaging information on the lower surface are collected, and the upper surface chroma imaging information and the lower surface chroma imaging information are respectively collected by the chroma image signal collector, and then analyzed by software pre-programmed by the industrial computer 8. And processing images). That is, the image gray scale and chromaticity features are used to identify the defect point and the defect depth d is greater than 20% of the thickness ⁇ , and the area exceeds 2% of the board surface.
  • the upper surface automatic marking device 10 and the lower surface automatic marking device 11 are activated to identify the defect points, the defect point identification and the classification result are input into the self-learning defect point chromaticity and grayscale feature database 12;
  • the process is implemented by loading a pre-programmed driver and image feature database through an industrial computer.
  • the software system flow is as follows - the computer starts running, and the system self-test is performed according to the parameters initialized by the system parameters;
  • Illuminating and collecting images driving the light source to illuminate both sides of the steel plate, and simultaneously imaging the camera to collect grayscale and chromaticity imaging information of the upper and lower surface defects of the steel plate;
  • the marking equipment marks the surface defect and the unqualified area ratio of the steel plate and the position on the X and Y coordinates.
  • the color image of the steel plate defect is first taken online, and the weight is set according to the RGB value of the color image, that is, the color image formed by the steel plate is converted into a gray image according to the color of the steel plate, so that the contrast is good.
  • Original grayscale image Then the gradient operator is used to detect the edge of the defect image, and the gradient image obtained by the edge detection is gray-scaled to improve the contrast of the gradient image. Then the optimal threshold is used to segment the defect. Finally, the noise is filtered. Automatic segmentation and positioning of defects. The features such as defect size, shape, color and density are obtained, and the defects are classified according to the characteristics of different defects, and the characteristic data of the steel plate defect characteristics database is established by using the characteristic data.
  • the industrial computer 8 passes the image collectors 6 and ⁇ to the high-speed color imaging systems 13, 14, 15, 16, 17, 18, respectively. 19, the imaging command is issued, and at the same time, the light source is driven to illuminate both sides of the steel plate, the camera is quickly imaged, and the color image is formed on the surface of the steel plate by the integrated image system of 6 and 7 inches.
  • the weight is set according to the RGB value of the color image, that is, the color image formed by the steel plate is converted into a grayscale image according to the color of the steel plate by the online variable weight color conversion method, thereby obtaining a raw grayscale image with good contrast.
  • the gradient operator is used to detect the edge of the defect image, and the gradient image obtained by the edge detection is gray-scaled to improve the contrast of the gradient image. Then Fisher's criterion is used to find the optimal threshold and the defect is segmented. Finally, the mathematical morphology is applied. The noise is filtered out, and the automatic segmentation and positioning of defects is realized. Obtain the characteristics of defect size, shape, color and density, and query the defect feature database according to the characteristic data to obtain the defect type of the steel plate, and alarm, mark and record (and count) according to the type of the defect.
  • the measurement process is as follows: (1) Defect imaging
  • the weight is set according to the RGB value of the on-line photographing of the defective color image of the steel sheet, that is, the color image formed by the steel sheet is converted into a gray scale image according to the color of the steel sheet by the online variable weight color conversion method, thereby obtaining a raw gray scale image with good contrast.
  • Figure 4 is a surface diagram of a steel plate with a scratch defect
  • Figure 5 is a histogram of the figure. It can be seen from Fig. 4 and Fig. 5 that the surface image of the steel sheet has the following characteristics: (1) the area of the surface defect of the steel sheet is small, and the histogram does not exhibit a bimodal form; (2) in the image of the surface of the steel sheet, the background The gray value is not uniform; (3) The gray level of the defect is not much different from the background, but the gradient of the background is small, and the gradient of the defect is large. Using the edge information of the image, the boundary of the image defect target can be quickly and accurately found, and the segmentation of the defect can be realized.
  • Figure 6 shows the results of the Sobel gradient operator processing of Figure 4.
  • grayscale stretching can selectively stretch the grayscale interval of interest to improve the contrast of the image.
  • the function expression for grayscale stretching is as follows: c 2
  • / represents the gray value of the image before the transformation
  • / represents the gray value of the transformed image
  • c, c 2 are segmentation points, which are selected according to the image features.
  • the gradient diagram shown in Figure 7 is compared to the histogram before and after grayscale stretching. As shown in Fig. 7, it can be seen that the gray scale dynamic range of the image after stretching is expanded.
  • the gradient map obtained by edge detection can be divided into two types: background gradient and defect gradient.
  • the gradient threshold can separate the two types, and the more accurate the threshold is, the better the separation is, so find an optimal gradient threshold.
  • the background gradient and the defect gradient are optimally separated, thereby segmenting the defects.
  • the threshold segmentation algorithm separates the defects and background, but there are still some noises that need to be denoised. When the image is filtered out, it should have the characteristics of unblurred image boundaries, preserve image geometry and geometric features.
  • pattern matching is performed on the defect image, that is, the defect is identified and classified, and if there is no defect, the next image is prepared, and when there is a defect, the system reaches the standard reaching device. Marking, and simultaneously alerting, and counting the number, location, and type of defects.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Textile Engineering (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

Un procédé d'examen automatique en ligne de détection de défauts de surface de l'acier pendant le prétraitement d'une tôle d'acier consiste à: éclairer les surfaces supérieure et inférieure de l'acier avec un module éclairage; imager les deux surfaces de l'acier avec un moyen d'imagerie pour créer une image; analyser l'image avec un processeur d'image; et déterminer s'il existe une quelconque surface défectueuse d'après l'information sur le niveau de gris de l'image. En cas de détection d'un défaut, le moyen de marquage s'enclenche et marque le point défectueux.
PCT/CN2006/002403 2005-12-01 2006-09-14 Procede d'examen automatique en ligne pour des detections de defauts de surface d'acier pendant le pretraitement de toles d'acier WO2007062563A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200510047897.4 2005-12-01
CNB2005100478974A CN100485371C (zh) 2005-12-01 2005-12-01 钢板预处理过程中的上、下表面缺陷在线自动检测方法

Publications (1)

Publication Number Publication Date
WO2007062563A1 true WO2007062563A1 (fr) 2007-06-07

Family

ID=36787992

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2006/002403 WO2007062563A1 (fr) 2005-12-01 2006-09-14 Procede d'examen automatique en ligne pour des detections de defauts de surface d'acier pendant le pretraitement de toles d'acier

Country Status (2)

Country Link
CN (1) CN100485371C (fr)
WO (1) WO2007062563A1 (fr)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107486415A (zh) * 2017-08-09 2017-12-19 中国计量大学 基于机器视觉的篾片缺陷在线检测系统及检测方法
CN108548822A (zh) * 2018-06-21 2018-09-18 无锡旭锠智能科技有限公司 一种宽幅连续表面缺陷视觉检测系统
CN109949292A (zh) * 2019-03-20 2019-06-28 湖南镭目科技有限公司 一种带钢表面缺陷的检测方法及装置
CN112697803A (zh) * 2020-12-16 2021-04-23 沈阳建筑大学 一种基于机器视觉的板带钢表面缺陷检测方法及装置
CN113160112A (zh) * 2020-12-30 2021-07-23 苏州特文思达科技有限公司 一种基于机器学习的钢格板产品智能识别检测方法
CN113976669A (zh) * 2021-10-26 2022-01-28 首钢京唐钢铁联合有限责任公司 一种热轧带钢表面缺陷的修复方法
CN114152622A (zh) * 2021-11-30 2022-03-08 南方电网物资有限公司 一种电力物资入库检测系统
CN115330781A (zh) * 2022-10-13 2022-11-11 启东谷诚不锈钢制品有限公司 一种钢板缺陷识别方法
CN116645367A (zh) * 2023-07-27 2023-08-25 山东昌啸商贸有限公司 一种高端制造用的钢板切割质量检测方法
CN117893542A (zh) * 2024-03-18 2024-04-16 西安华联电力电缆有限公司 一种基于图像处理技术的绝缘电缆生产方法及系统

Families Citing this family (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100440083C (zh) * 2006-06-27 2008-12-03 宝山钢铁股份有限公司 轧制大板的在线优化剪切方法
CN102192909A (zh) * 2010-03-12 2011-09-21 深圳市先阳软件技术有限公司 一种电池极片的激光切割质量检测系统
CN103207185A (zh) * 2012-01-11 2013-07-17 宝山钢铁股份有限公司 钢卷端部质量检测系统及方法
CN102565075A (zh) * 2012-01-13 2012-07-11 博罗承创精密工业有限公司 一种led支架自动检测装置
CN102721699A (zh) * 2012-06-06 2012-10-10 朱保凌 一种高清探头无损检测系统
CN103529043B (zh) * 2012-07-06 2015-10-28 中国钢铁股份有限公司 钢胚缺陷光学检测系统及其检测方法
CN102841101B (zh) * 2012-08-16 2014-08-20 南京协力电子科技集团有限公司 Pcb线路板直立式双面ccd线扫描自动化检测平台
CN103077526B (zh) * 2013-02-01 2016-05-11 苏州华兴致远电子科技有限公司 具有深度检测功能的列车异常检测方法及系统
FR3005042B1 (fr) * 2013-04-26 2016-01-01 Snecma Machine a tisser ou enrouler une texture fibreuse permettant un controle d'anomalies par analyse d'images
CN103662955B (zh) * 2013-05-11 2016-01-20 湖北南天无纺布制品有限公司 汽车内饰件用无纺布裁切机及裁切方法
CN103586166A (zh) * 2013-09-26 2014-02-19 中航锂电（洛阳）有限公司 一种涂布生产线及其在线自动标识装置
CN103575741A (zh) * 2013-10-29 2014-02-12 浚丰太阳能(江苏)有限公司 一种底面外观检查器
CN104897679A (zh) * 2014-03-05 2015-09-09 鞍钢股份有限公司 一种钢板表面缺陷的检查方法
CN104048973B (zh) * 2014-05-28 2016-08-17 苏州科技学院 卷状遮光料的漏光检测装置
CN104165842B (zh) * 2014-07-21 2016-08-10 四川祥益智能科技有限公司 一种在线光学检测仪
CN105784720A (zh) * 2014-12-16 2016-07-20 徐帆 布料缺陷的检测方法和系统
CN104476142B (zh) * 2014-12-19 2017-06-16 青岛维尔环保科技有限公司 滑道梁防变形制作工艺
CN105136810A (zh) * 2015-08-27 2015-12-09 张小磊 基于高强度照明的钢板缺陷检测平台
CN105092593A (zh) * 2015-08-27 2015-11-25 张小磊 一种基于高强度照明的钢板缺陷检测方法
CN105842253B (zh) * 2016-03-18 2019-07-23 奥士康科技股份有限公司 一种自动化pcb板光学检测方法
CN109342435B (zh) * 2016-08-29 2021-02-05 李建勋 全自动工件表面质量检测装置
CN106383124A (zh) * 2016-08-30 2017-02-08 四川石棉华瑞电子有限公司 电容铝箔表面缺陷视觉在线检查系统及方法
CN108508031A (zh) * 2017-02-28 2018-09-07 上海微电子装备（集团）股份有限公司 一种双面检测装置及检测方法
CN106872488A (zh) * 2017-04-20 2017-06-20 广东振华科技股份有限公司 一种快速大面积透明基片双表面缺陷视觉检测方法及装置
CN110308256B (zh) * 2018-03-20 2022-07-05 宝山钢铁股份有限公司 钢板下表周期缺陷测量装置
TWI721427B (zh) * 2018-05-22 2021-03-11 荷蘭商耐克創新有限合夥公司 皮料檢測設備及偵測皮坯的上表面及下表面兩者上的瑕疵的方法
CN109064454A (zh) * 2018-07-12 2018-12-21 上海蝶鱼智能科技有限公司 产品缺陷检测方法及系统
CN109358073A (zh) * 2018-11-12 2019-02-19 威海华菱光电股份有限公司 圆棒的检测装置
CN110899150B (zh) * 2019-12-23 2021-02-23 中国环境科学研究院 一种电解锌锰阴阳极表面物理缺陷智能识别的方法
CN111862028B (zh) * 2020-07-14 2021-04-09 南京林业大学 基于深度相机和深度学习的木材缺陷检测分选装置及方法
CN111982919A (zh) * 2020-08-25 2020-11-24 上海上创超导科技有限公司 涂层导体金属基带表面缺陷的动态监测装置及检测方法
CN112763503A (zh) * 2020-12-24 2021-05-07 山西大数据产业发展有限公司 基于机器视觉的高牌号冷轧硅钢孔洞检测系统
CN113077414B (zh) * 2021-03-01 2023-05-16 湖北工程学院 一种钢板表面缺陷检测方法及系统
CN113073276A (zh) * 2021-03-29 2021-07-06 江西耐乐铜业有限公司 一种基于二次退火的铜管在线退火系统及测试方法
CN113189111B (zh) * 2021-04-28 2022-11-01 西南交通大学 一种钢结构网架外观缺陷视觉检测系统及检测方法
CN113670941A (zh) * 2021-08-30 2021-11-19 宝钢湛江钢铁有限公司 一种多功能连退带钢表面氧化色缺陷在线定量评价的方法
CN116309375B (zh) * 2023-02-23 2023-10-24 南京林业大学 实木板材双面缺陷检测及智能加工坐标确定方法
CN117517348B (zh) * 2023-11-14 2024-05-14 四川领先微晶玻璃有限公司 基于微晶玻璃面板成品的表面缺陷检测系统

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4974261A (en) * 1988-11-15 1990-11-27 Matsushita Electric Works, Ltd. Optical surface inspection method
US5204911A (en) * 1991-05-29 1993-04-20 Nira Schwartz Inspection method using unique templates and histogram analysis
US5274713A (en) * 1991-09-23 1993-12-28 Industrial Technology Research Institute Real-time apparatus for detecting surface defects on objects
JPH07229832A (ja) * 1994-02-16 1995-08-29 Nkk Corp 表面検査方法及び装置
US5481619A (en) * 1992-07-07 1996-01-02 Schwartz; Nira Inspection method using templates images, unique histogram analysis, and multiple gray levels correlated to addresses of template images
JPH0882604A (ja) * 1994-09-12 1996-03-26 Nippon Steel Corp 鋼板表面欠陥検査方法
JPH08189904A (ja) * 1995-01-06 1996-07-23 Kawasaki Steel Corp 表面欠陥検出装置
JPH10232206A (ja) * 1997-02-19 1998-09-02 Hitachi Ltd 金属表面欠陥検出方法及び装置
JP2004045356A (ja) * 2002-05-20 2004-02-12 Jfe Steel Kk 表面欠陥検出方法
US20040141640A1 (en) * 2003-01-08 2004-07-22 Byoung-Ho Lee Method for apparatus for detecting defects on a wafer
WO2004107268A1 (fr) * 2003-05-30 2004-12-09 Zen Voce Manufacturing Pte Ltd Procede et systeme de detection de non uniformite de la surface superieure d'attaches

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4974261A (en) * 1988-11-15 1990-11-27 Matsushita Electric Works, Ltd. Optical surface inspection method
US5204911A (en) * 1991-05-29 1993-04-20 Nira Schwartz Inspection method using unique templates and histogram analysis
US5274713A (en) * 1991-09-23 1993-12-28 Industrial Technology Research Institute Real-time apparatus for detecting surface defects on objects
US5481619A (en) * 1992-07-07 1996-01-02 Schwartz; Nira Inspection method using templates images, unique histogram analysis, and multiple gray levels correlated to addresses of template images
JPH07229832A (ja) * 1994-02-16 1995-08-29 Nkk Corp 表面検査方法及び装置
JPH0882604A (ja) * 1994-09-12 1996-03-26 Nippon Steel Corp 鋼板表面欠陥検査方法
JPH08189904A (ja) * 1995-01-06 1996-07-23 Kawasaki Steel Corp 表面欠陥検出装置
JPH10232206A (ja) * 1997-02-19 1998-09-02 Hitachi Ltd 金属表面欠陥検出方法及び装置
JP2004045356A (ja) * 2002-05-20 2004-02-12 Jfe Steel Kk 表面欠陥検出方法
US20040141640A1 (en) * 2003-01-08 2004-07-22 Byoung-Ho Lee Method for apparatus for detecting defects on a wafer
WO2004107268A1 (fr) * 2003-05-30 2004-12-09 Zen Voce Manufacturing Pte Ltd Procede et systeme de detection de non uniformite de la surface superieure d'attaches

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107486415B (zh) * 2017-08-09 2023-10-03 中国计量大学 基于机器视觉的篾片缺陷在线检测系统及检测方法
CN107486415A (zh) * 2017-08-09 2017-12-19 中国计量大学 基于机器视觉的篾片缺陷在线检测系统及检测方法
CN108548822A (zh) * 2018-06-21 2018-09-18 无锡旭锠智能科技有限公司 一种宽幅连续表面缺陷视觉检测系统
CN109949292A (zh) * 2019-03-20 2019-06-28 湖南镭目科技有限公司 一种带钢表面缺陷的检测方法及装置
CN109949292B (zh) * 2019-03-20 2023-11-14 湖南镭目科技有限公司 一种带钢表面缺陷的检测方法及装置
CN112697803A (zh) * 2020-12-16 2021-04-23 沈阳建筑大学 一种基于机器视觉的板带钢表面缺陷检测方法及装置
CN113160112A (zh) * 2020-12-30 2021-07-23 苏州特文思达科技有限公司 一种基于机器学习的钢格板产品智能识别检测方法
CN113160112B (zh) * 2020-12-30 2024-02-02 苏州特文思达科技有限公司 基于机器学习的钢格板产品智能识别检测方法
CN113976669A (zh) * 2021-10-26 2022-01-28 首钢京唐钢铁联合有限责任公司 一种热轧带钢表面缺陷的修复方法
CN114152622A (zh) * 2021-11-30 2022-03-08 南方电网物资有限公司 一种电力物资入库检测系统
CN115330781A (zh) * 2022-10-13 2022-11-11 启东谷诚不锈钢制品有限公司 一种钢板缺陷识别方法
CN116645367A (zh) * 2023-07-27 2023-08-25 山东昌啸商贸有限公司 一种高端制造用的钢板切割质量检测方法
CN116645367B (zh) * 2023-07-27 2023-12-01 山东昌啸商贸有限公司 一种高端制造用的钢板切割质量检测方法
CN117893542A (zh) * 2024-03-18 2024-04-16 西安华联电力电缆有限公司 一种基于图像处理技术的绝缘电缆生产方法及系统

Also Published As

Publication number Publication date
CN1789990A (zh) 2006-06-21
CN100485371C (zh) 2009-05-06

Similar Documents

Publication Publication Date Title
WO2007062563A1 (fr) Procede d'examen automatique en ligne pour des detections de defauts de surface d'acier pendant le pretraitement de toles d'acier
US7570794B2 (en) System and method for evaluating a machined surface of a cast metal component
US8050486B2 (en) System and method for identifying a feature of a workpiece
CN109596530B (zh) 光学表面麻点和灰尘缺陷分类的暗场偏振成像方法
CN106651857B (zh) 一种印刷电路板贴片缺陷检测方法
TWI669519B (zh) 電路板瑕疵篩選方法及其裝置與電腦可讀取記錄媒體
JP3938227B2 (ja) 異物検査方法および装置
JPH1163959A (ja) 表面検査装置
CN115601359A (zh) 一种焊缝检测方法及装置
CN111879789A (zh) 金属表面缺陷检测方法及系统
JP2010054502A (ja) 熱交換器のチューブ抽出検査方法及び装置
CN111833350A (zh) 机器视觉检测方法与系统
JP4652024B2 (ja) 表面検査方法及び装置
KR20190119801A (ko) 차량 헤드라이트 얼라인먼트 보정 및 분류 방법 및 이를 이용한 차량 헤드라이트 불량검사 방법
TWI777307B (zh) 在太陽能模組影像中以深度學習進行太陽能電池瑕疵辨識的方法、電腦程式及電腦可讀取媒體
CN115165920A (zh) 一种三维缺陷检测方法及检测设备
KR101782366B1 (ko) 스케일 보정을 통한 표본 정규화가 수행되는 학습기반의 비전검사 방법
TWI749714B (zh) 瑕疵檢測方法、瑕疵分類方法及其系統
JP4052733B2 (ja) パターン付きウエハの異物検査方法
TWM606740U (zh) 瑕疵檢測系統
JPH11101750A (ja) 異物検出方法
JP2010038650A (ja) 熱交換器のチューブ幅検査方法及び装置
JP2007183283A (ja) 異物検査方法および装置
TWI816549B (zh) 自動化瑕疵檢測方法
JPH07159333A (ja) 外観検査装置及び外観検査方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 06775662

Country of ref document: EP

Kind code of ref document: A1