CN106442556A - Device and method for detecting surface defects of perforated plate workpiece - Google Patents
Device and method for detecting surface defects of perforated plate workpiece Download PDFInfo
- Publication number
- CN106442556A CN106442556A CN201611007125.2A CN201611007125A CN106442556A CN 106442556 A CN106442556 A CN 106442556A CN 201611007125 A CN201611007125 A CN 201611007125A CN 106442556 A CN106442556 A CN 106442556A
- Authority
- CN
- China
- Prior art keywords
- image
- shaped
- threshold
- special
- value
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000001514 detection method Methods 0.000 claims abstract description 19
- 239000011159 matrix material Substances 0.000 claims description 33
- 238000001228 spectrum Methods 0.000 claims description 24
- 230000005540 biological transmission Effects 0.000 claims description 10
- 230000003595 spectral effect Effects 0.000 claims description 10
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008676 import Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000003709 image segmentation Methods 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 238000001914 filtration Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- 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
-
- 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
- G01N2021/8887—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 based on image processing techniques
Landscapes
- Chemical & Material Sciences (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Textile Engineering (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
Description
技术领域technical field
本发明涉及一种板状带孔工件表面缺陷检测装置和方法，属于视觉检测领域。The invention relates to a surface defect detection device and method of a plate-shaped workpiece with holes, belonging to the field of visual detection.
背景技术Background technique
板状带孔工件，近年来被广泛应用于机床设备加工、医疗器械制造、车辆装置方面。目前大多数生产厂家仍采用人工方式寻找板状带孔工件表面缺陷，不仅对人眼伤害大，而且效率低、成本高，不能准确发现工件表面缺陷的具体位置，而且容易使人产生视觉误差，导致工件质量下降，出现漏检、误检，大大降低了产品价格和市场竞争力。Plate-shaped workpieces with holes have been widely used in machine tool processing, medical equipment manufacturing, and vehicle installations in recent years. At present, most manufacturers still use manual methods to find surface defects of plate-shaped workpieces with holes, which is not only harmful to human eyes, but also has low efficiency and high cost. It cannot accurately find the specific position of surface defects on the workpiece, and it is easy to cause visual errors This leads to a decline in the quality of the workpiece, missed inspections, and false inspections, which greatly reduces product prices and market competitiveness.
曾有生产厂家通过技术改进，采用激光扫描检测，以多点排列的点光源照射工件的表面，通过观察到达接收板的点光源的缺省程度来判断工件表面的完整程度。该检测方法不但动作复杂，需要提前排列激光发生器的发光点位置，并且激光发生器的成本及后期维护费用也比较高，增加了探伤检测的成本。There have been manufacturers who have adopted laser scanning detection through technical improvement, and irradiated the surface of the workpiece with multi-point light sources, and judged the integrity of the workpiece surface by observing the default degree of the point light sources reaching the receiving plate. This detection method is not only complicated in action, but also needs to arrange the position of the light-emitting point of the laser generator in advance, and the cost of the laser generator and the later maintenance cost are also relatively high, which increases the cost of flaw detection.
发明内容Contents of the invention
针对上述现有技术，本发明提供了一种板状带孔工件表面缺陷检测装置和方法，用以解决上述板状带孔工件表面缺陷检测所存在的技术问题。In view of the above prior art, the present invention provides a device and method for detecting surface defects of a plate-shaped workpiece with holes to solve the technical problems existing in the detection of surface defects of the above-mentioned plate-shaped workpiece with holes.
本发明一种板状带孔工件表面缺陷检测装置予以实现的技术方案是：该装置包括传送系统、特形照明光源、工业CCD图像传感器、图像采集卡及处理器；所述的传送系统用于水平传送待检测板状带孔工件，并以匀速直线形式进行传送；所述的特形照明光源用于提供均匀照明光源，并且特形照明光源能够覆盖整个待检测板状带孔工件所在区域；所述的工业CCD图像传感器用于将光学信号转换为电信号，完成图像的采集部分；所述的图像采集卡用于接收从摄像头采集的电信号，并将收集到的模拟信号经过A/D转换，对图像信息进行存储和处理，并将数据信息传输给处理器；所述的处理器是在编程环境中实现相应代码的运行，计算、标记出缺陷位置并直观显示出来；The technical solution of the present invention for detecting a surface defect of a plate-shaped workpiece with holes is: the device includes a transmission system, a special-shaped lighting source, an industrial CCD image sensor, an image acquisition card and a processor; the transmission system is used for The plate-shaped workpiece with holes to be detected is conveyed horizontally, and conveyed in a straight line at a constant speed; the special-shaped lighting source is used to provide a uniform lighting source, and the special-shaped lighting source can cover the entire area where the plate-shaped workpiece with holes to be detected is located; The industrial CCD image sensor is used to convert the optical signal into an electrical signal to complete the image acquisition part; the image acquisition card is used to receive the electrical signal collected from the camera, and pass the collected analog signal through the A/D Converting, storing and processing the image information, and transmitting the data information to the processor; the processor implements the operation of the corresponding code in the programming environment, calculates, marks the defect position and visually displays it;
所述的待检测板状带孔工件水平置于所述的传送系统中；所述的特形照明光源为顶部是弧形、水平切面是矩形的棚状结构光源，并且特形照明光源能够覆盖整个所述的待检测板状带孔工件所在区域；所述的特形照明光源设置在所述的工业CCD图像传感器的镜头四周，并与所述的工业CCD图像传感器连接固定；所述的特形照明光源的几何模型的数学表达式为：The plate-shaped workpiece with holes to be inspected is placed horizontally in the transmission system; the special-shaped lighting source is a shed-shaped light source with an arc-shaped top and a rectangular horizontal section, and the special-shaped lighting source can cover The entire area where the plate-shaped workpiece with holes to be detected is located; the special-shaped lighting source is arranged around the lens of the industrial CCD image sensor, and is connected and fixed with the industrial CCD image sensor; the special The mathematical expression of the geometric model of the shaped lighting source is:
(1) (1)
其中，、为未知数系数，、、为边界值；所述的工业CCD图像传感器包括：摄像机主体、镜头及C接口；所述的摄像机主体采用工业CCD摄像机；所述的摄像机主体与镜头通过C接口连接；所述的镜头设置在所述的特形照明光源内侧，并垂直于所述的待检测板状带孔工件；所述的图像采集卡为图像采集部分与图像处理部分的接口。in, , is the unknown coefficient, , , is boundary value; described industrial CCD image sensor comprises: camera main body, lens and C interface; Described camera main body adopts industrial CCD camera; Described camera main body and lens are connected by C interface; Described lens is arranged on the The inner side of the special-shaped lighting source, and perpendicular to the plate-shaped workpiece with holes to be inspected; the image acquisition card is the interface between the image acquisition part and the image processing part.
本发明提出的一种板状带孔工件表面缺陷检测方法，是利用上述一种板状带孔工件表面缺陷检测装置，并按照以下步骤：A method for detecting surface defects of a plate-shaped workpiece with holes proposed by the present invention is to use the above-mentioned device for detecting surface defects of a plate-shaped workpiece with holes, and follow the steps below:
所述的待检测板状带孔工件水平置于所述的传送系统中，并以直线的形式匀速地从镜头下方移过，镜头实时采集完整的图像信息；所述的工业CCD图像传感器将光学信号转换为电信号，再经过所述的图像采集卡接收从摄像头采集到的电信号，并将采集到的模拟信号经过A/D转换，对图像信息进行存储和处理，并由图像采集卡将数据信息传输给所述的处理器；在处理器中，首先通过编程软件对图像进行预处理，改善图像的视觉效果和清晰度，包含直方图均衡化处理、归一化处理、中值滤波，再根据图像特征对图像进行二值化处理；接着使用边缘检测技术，在抑噪的同时用边缘点勾画出各个对象的轮廓，分析图像某些需要识别的目标；然后通过图像分割技术，对比图像中显著值的异常位置进行区域标记，最后提取图像特征，并直观发现缺陷所在位置。The plate-shaped workpiece with holes to be detected is placed horizontally in the transmission system, and moves under the lens at a constant speed in a straight line, and the lens collects complete image information in real time; the industrial CCD image sensor combines optical The signal is converted into an electrical signal, and then the electrical signal collected from the camera is received through the image acquisition card, and the analog signal collected is converted through A/D, and the image information is stored and processed, and the image information is stored and processed by the image acquisition card. The data information is transmitted to the processor; in the processor, the image is first preprocessed by programming software to improve the visual effect and clarity of the image, including histogram equalization processing, normalization processing, median filtering, Then binarize the image according to the image features; then use edge detection technology to outline the outline of each object with edge points while suppressing noise, and analyze some objects that need to be identified in the image; then use image segmentation technology to compare the images The abnormal position of the significant value in the middle is marked, and finally the image features are extracted, and the position of the defect is intuitively found.
步骤一、谱剩余算法计算：Step 1. Spectral residual algorithm calculation:
首先对输入的灰度图像进行二维离散傅里叶变换，将图像从空间域转入频率域:，其中，为灰度图像空间域坐标，为灰度图像频率域坐标；First, two-dimensional discrete Fourier transform is performed on the input grayscale image, and the image is transferred from the spatial domain to the frequency domain: ,in, is the spatial domain coordinates of the grayscale image, is the grayscale image frequency domain coordinates;
(2) (2)
其中，为该点傅里叶频谱值。再求幅度谱和相位谱：in, is the Fourier spectrum value of this point. Then find the magnitude spectrum and phase spectrum:
(3) (3)
(4) (4)
对幅度谱取对数，得到其幅值的Log谱：Take the logarithm of the magnitude spectrum to get the Log spectrum of its magnitude:
(5) (5)
然后对Log谱进行平滑滤波，获取Log谱的大概形状：Then smooth and filter the Log spectrum to obtain the approximate shape of the Log spectrum:
(6) (6)
其中，是一个的平滑滤波器，为平滑滤波器的空域带宽。求取两者的差值，得到谱残差：in, Is a smoothing filter, is the spatial bandwidth of the smoothing filter. Find the difference between the two to get the spectral residual:
(7) (7)
对谱残差和相位谱进行二维傅里叶逆变换，得pair spectral residual and phase spectrum Carrying out two-dimensional inverse Fourier transform, we get
(8) (8)
其中，表示灰度图像中每点坐标的显著值。in, Indicates the saliency value of each point coordinate in the grayscale image.
步骤二、阈值的设定：Step 2. Threshold setting:
本发明采用两种方法设置阈值，并分别将阈值与同一幅图像中各像素点的显著值对比，将显著值大于等于阈值的像素点标记为“1”，记为目标区域；显著值小于阈值的像素点标记为“0”，记为背景区域。The present invention adopts two methods to set the threshold , and respectively compare the threshold with the salient value of each pixel in the same image, mark the pixel with a salient value greater than or equal to the threshold as "1" and record it as the target area; mark the pixel with a salient value smaller than the threshold as "0" , recorded as the background area.
根据自适应阈值算法求解阈值。将设为给定图像的平均显著值：Solve the threshold according to the adaptive threshold algorithm . Will Set to mean saliency for a given image:
(9) (9)
其中，、对应图像的长和宽。将获取的图像的每处显著值与自适应阈值比较，将大于等于阈值的像素点标记为“1”，小于阈值的像素点标记为“0”， 并把该幅图像中的所有像素点集合用由0、1组成的行列矩阵表示。in, , corresponds to the height and width of the image. Combine each salient value of the acquired image with an adaptive threshold For comparison, the pixels greater than or equal to the threshold are marked as "1", and the pixels less than the threshold are marked as "0", and all the pixel points in the image are set by 0, 1 Row column matrix express.
根据大律法求解阈值。将该幅图像显著值范围记为，(为最大显著值)。预先设置一阈值将上述图像显著值分为两类：,，并将与分别记为目标与背景，两者的类间方差为：Solving Thresholds According to the Big Law . The significant value range of this image is recorded as , ( is the maximum significant value). Preset a threshold The above image saliency values are divided into two categories: , , and will and They are respectively recorded as the target and the background, and the variance between the two classes is:
(10) (10)
(11) (11)
(12) (12)
其中，表示图像中显著值低于的像素个数，表示图像中显著值高于等于的像素个数，为低于的总像素的平均显著值，为高于等于的总像素的平均显著值；in, Indicates that the significant value in the image is lower than the number of pixels, Indicates that the significant value in the image is higher than or equal to the number of pixels, for less than The average saliency value of the total pixels of for higher than or equal to The average saliency value of the total pixels of ;
使得值最大的值就是所需的阈值，即。再将与图像的每处显著值比较，将大于等于阈值的像素点标记为“1”，小于阈值的像素点标记为“0”， 并把该幅图像中的所有像素点集合用由0、1组成的行列矩阵表示。make the largest value value is the desired threshold ,Right now . then Compared with each significant value of the image, mark the pixels greater than or equal to the threshold as "1", and mark the pixels less than the threshold as "0", and use the set of all pixels in the image to be composed of 0 and 1 of Row column matrix express.
步骤三、标记缺陷Step 3. Mark defects
将矩阵与矩阵对应元素相乘，得新矩阵。即：。由矩阵点乘公式可知，矩阵也是由元素0、1构成，元素1表示该像素点显著值均大于等于两阈值，即目标图像的重合位置。再按照从左往右、从上往下的顺序依次寻找三个矩阵、、中元素为1的连通区，分别将每个矩阵的连通区记为、、 。选取矩阵，对矩阵中所有元素求和，记和为；对矩阵中所有元素求和，记和为。引入函数，令 (13)the matrix with matrix Multiply the corresponding elements to get a new matrix . which is: . According to the matrix dot product formula, the matrix It is also composed of elements 0 and 1, and element 1 indicates that the salient values of the pixel are greater than or equal to the two thresholds , which is the overlapping position of the target image. Then search for three matrices in order from left to right and from top to bottom , , In the connected area where the element is 1, the connected area of each matrix is recorded as , , . selection matrix , for the matrix The sum of all elements in the sum is recorded as ; pair matrix The sum of all elements in the sum is recorded as . import function ,make (13)
若，则认为在r处为缺陷位置。若小于，则认为是误差检测，工件此处无缺陷。like , then it is considered that the position at r is the defect position. if less than , it is considered as an error detection, and the workpiece has no defects here.
与现有技术相比，本发明的有益效果是：Compared with prior art, the beneficial effect of the present invention is:
本发明提供的一种板状带孔工件表面缺陷检测装置和方法，采用特形照明光源取代现有的点光源，能够均匀照射，不存在光差现象，解决了待检测板状带孔工件表面检测受照明程度影响导致采集图像不清晰的问题。本发明对于待检测板状带孔工件图像的采集方式不存在同步时差，能够实时显示待检测板状带孔工件表面缺陷位置，软硬件互换度高，适用于动态目标物的视觉成像，且图像精度高、色彩还原度高。The present invention provides a surface defect detection device and method for a plate-shaped workpiece with holes, which uses a special-shaped lighting source to replace the existing point light source, can illuminate evenly, and has no light difference phenomenon, and solves the problem of the surface defect of the plate-shaped workpiece with holes to be inspected. Detect the problem that the acquired image is not clear due to the influence of lighting level. The present invention has no synchronous time difference for the acquisition method of the image of the plate-shaped workpiece with holes to be detected, can display the surface defect position of the plate-shaped workpiece with holes to be detected in real time, has a high degree of interchangeability of software and hardware, and is suitable for visual imaging of dynamic targets, and High image precision and high color reproduction.
附图说明Description of drawings
图1为本发明提供的一种板状带孔工件表面缺陷检测装置结构图；Fig. 1 is a structural diagram of a surface defect detection device for a plate-shaped workpiece with holes provided by the present invention;
图2为本发明提供的特形照明光源结构示意图；Fig. 2 is a structural schematic diagram of a special-shaped lighting source provided by the present invention;
图3为本发明提供的一种板状带孔工件表面缺陷检测装置的检测系统流程图。Fig. 3 is a flow chart of a detection system of a surface defect detection device for a plate-shaped workpiece with holes provided by the present invention.
图中：1-传送系统，2-待检测板状带孔工件，3-特形照明光源，4-工业CCD图像传感器，5-图像采集卡，6-处理器。In the figure: 1-transmission system, 2-plate-shaped workpiece with holes to be inspected, 3-special-shaped lighting source, 4-industrial CCD image sensor, 5-image acquisition card, 6-processor.
具体实施方式detailed description
下面结合具体实施方式对本发明作进一步详细地描述。The present invention will be further described in detail below in combination with specific embodiments.
如图1所示，本发明一种板状带孔工件表面缺陷检测装置，包括传送系统1、特形照明光源3、工业CCD图像传感器4、图像采集卡5及处理器6。所述的传送系统1用于水平传送所述的待检测板状带孔工件2，并以匀速直线形式进行传送；所述的特形照明光源3用于提供均匀照明光源，并且特形照明光源3能够覆盖整个待检测板状带孔工件2所在区域；所述的工业CCD图像传感器4用于将光学信号转换为电信号，完成图像的采集部分；所述的图像采集卡5用于接收从摄像头采集的电信号，并将收集到的模拟信号经过A/D转换，对图像信息进行存储和处理，并将数据信息传输给所述的处理器6；所述的处理器6是在编程环境中实现相应代码的运行，计算、标记出缺陷位置并直观显示出来。As shown in FIG. 1 , a device for detecting surface defects of a plate-shaped workpiece with holes in the present invention includes a transmission system 1 , a special-shaped lighting source 3 , an industrial CCD image sensor 4 , an image acquisition card 5 and a processor 6 . The conveying system 1 is used to horizontally convey the plate-shaped workpiece 2 with holes to be inspected and conveyed in a straight line at a constant speed; the special-shaped lighting source 3 is used to provide a uniform lighting source, and the special-shaped lighting source 3 can cover the area where the entire plate-like workpiece with holes to be detected is located; the industrial CCD image sensor 4 is used to convert optical signals into electrical signals to complete the image acquisition part; the image acquisition card 5 is used to receive from The electrical signal collected by the camera, and the collected analog signal is A/D converted, the image information is stored and processed, and the data information is transmitted to the processor 6; the processor 6 is programmed in the environment Realize the operation of the corresponding code in the program, calculate and mark the defect position and display it visually.
所述的待检测板状带孔工件2水平置于所述的传送系统1中；所述的特形照明光源3为顶部是弧形、水平切面是矩形的棚状结构光源，并且特形照明光源3能够覆盖整个所述的待检测板状带孔工件2所在区域；所述的特形照明光源3设置在所述的工业CCD图像传感器4的镜头四周，并与所述的工业CCD图像传感器4连接固定；所述的特形照明光源3的几何模型的数学表达式为：The plate-shaped workpiece 2 to be inspected is placed horizontally in the conveying system 1; the special-shaped lighting source 3 is a shed-shaped light source with an arc-shaped top and a rectangular horizontal section, and the special-shaped lighting The light source 3 can cover the entire area where the plate-shaped workpiece 2 to be detected is located; the special-shaped illumination light source 3 is arranged around the lens of the industrial CCD image sensor 4, and is connected with the industrial CCD image sensor 4. The connection is fixed; the mathematical expression of the geometric model of the special-shaped lighting source 3 is:
(1) (1)
其中，、为未知数系数，、、为边界值；所述的工业CCD图像传感器4包括摄像机主体、镜头及C接口；所述的摄像机主体采用工业CCD摄像机；所述的摄像机主体与镜头之间通过C接口连接；所述的镜头垂直于所述的待检测板状带孔工件2；所述的图像采集卡5为图像采集部分与图像处理部分的接口。in, , is the unknown coefficient, , , It is boundary value; Described industrial CCD image sensor 4 comprises camera main body, lens and C interface; Described camera main body adopts industrial CCD camera; Described camera main body and lens are connected by C interface; Described camera lens is vertical For the plate-shaped workpiece 2 to be inspected with holes; the image acquisition card 5 is the interface between the image acquisition part and the image processing part.
本发明提出的一种板状带孔工件表面缺陷检测方法，是利用上述一种板状带孔工件表面缺陷检测装置，并按照以下步骤：A method for detecting surface defects of a plate-shaped workpiece with holes proposed by the present invention is to use the above-mentioned device for detecting surface defects of a plate-shaped workpiece with holes, and follow the steps below:
所述的待检测板状带孔工件2水平置于所述的传送系统1中，并以直线的形式匀速地从镜头下方移过，镜头实时采集完整的图像信息；所述的工业CCD图像传感器4将光学信号转换为电信号，再经过所述的图像采集卡5接收从摄像头采集的电信号，并将采集到的模拟信号经过A/D转换，对图像信息进行存储和处理，并由图像采集卡5将数据信息传输给所述的处理器6；在处理器6中，首先通过编程软件对图像进行预处理，改善图像的视觉效果和清晰度，包含直方图均衡化处理、归一化处理、中值滤波，再根据图像特征对图像进行二值化处理；接着使用边缘检测技术，在抑噪的同时用边缘点勾画出各个对象的轮廓，分析图像某些需要识别的目标；然后通过图像分割技术，对比图像中显著值的异常位置进行区域标记，最后提取图像特征，并直观发现缺陷所在位置。The plate-shaped workpiece 2 to be detected is placed horizontally in the transmission system 1, and moves under the lens at a constant speed in a straight line, and the lens collects complete image information in real time; the industrial CCD image sensor 4 convert the optical signal into an electrical signal, then receive the electrical signal collected from the camera through the image acquisition card 5, and convert the collected analog signal through A/D conversion, store and process the image information, and use the image The acquisition card 5 transmits the data information to the processor 6; in the processor 6, the image is first preprocessed by programming software to improve the visual effect and clarity of the image, including histogram equalization processing, normalization processing, median filtering, and then binarize the image according to the image features; then use edge detection technology to outline the outline of each object with edge points while suppressing noise, and analyze some targets that need to be identified in the image; and then pass Image segmentation technology, compares the abnormal positions of significant values in the image to mark the area, and finally extracts image features, and intuitively finds the location of the defect.
如图3所示，一种板状带孔工件表面缺陷检测装置的检测流程依次为光源照射、图像采集传输、图像预处理、图像边缘检测、图像分割及图像识别，最终标记缺陷。As shown in Figure 3, the detection process of a surface defect detection device for plate-shaped workpieces with holes includes light source irradiation, image acquisition and transmission, image preprocessing, image edge detection, image segmentation and image recognition, and finally marks defects.
光源照射采用所述的特形照明光源3向所述的待检测板状带孔工件2投射均匀光。For light source irradiation, the special-shaped illumination source 3 is used to project uniform light to the plate-shaped workpiece 2 to be inspected with holes.
图像采集传输为所述的镜头实时采集完整的图像数据，通过所述的工业CCD图像传感器4将光学信号转换为电信号，再经过所述的图像采集卡5接收从摄像头采集到的电信号，并将收集到的模拟信号经过A/D转换，对图像信息进行采集存储和传输。The image acquisition transmission is the real-time collection of complete image data by the lens, the optical signal is converted into an electrical signal by the industrial CCD image sensor 4, and then the electrical signal collected from the camera is received by the image acquisition card 5, And the collected analog signal is converted by A/D, and the image information is collected, stored and transmitted.
图像预处理为所述的处理器6首先对图像建立灰度直方图，直观发现图像中像素亮度的分布情况；再对直方图进行均衡化、归一化处理，使图像的灰度均匀分布、增大反差，图像细节更加清晰；接着进行中值滤波，对图像滤除噪声，保护信号的细节信息，并保护图像边缘；最后对图像进行二值化处理，利用图像中的显著值差异，对比设定的阈值，将每个像素点归于一个区域，把目标区域标记为“1”，背景区域标记为“0”，从而一幅灰度图像变为二值图像。Image preprocessing is that described processor 6 first establishes a grayscale histogram to the image, and intuitively finds the distribution of pixel brightness in the image; then the histogram is equalized and normalized, so that the grayscale of the image is evenly distributed, Increase the contrast to make the details of the image clearer; then perform median filtering to filter out noise from the image, protect the details of the signal, and protect the edges of the image; finally, binarize the image, using the significant value difference in the image to compare The set threshold value assigns each pixel to an area, marks the target area as "1", and marks the background area as "0", so that a grayscale image becomes a binary image.
图像边缘检测为用边缘点勾画出各个对象的轮廓，分析出图像中需要识别的目标，即突出图像的边缘以提取图像特征。Image edge detection is to use edge points to outline the outline of each object, and analyze the target that needs to be recognized in the image, that is, to highlight the edge of the image to extract image features.
图像分割为对标识边缘检测后图像中亮度不同的点进行分割，将其划分为若干不重叠区域。Image segmentation is to segment the points with different brightness in the image after mark edge detection, and divide them into several non-overlapping regions.
图像识别为先采用谱剩余的显著目标检测法对所述的待检测板状带孔工件2进行检测，再使用综合比对法识别待检测板状带孔工件2表面缺陷位置。The image recognition is to first detect the plate-shaped hole-shaped workpiece 2 to be detected by using the spectral residual salient target detection method, and then use the comprehensive comparison method to identify the surface defect position of the plate-shaped hole-shaped workpiece 2 to be detected.
步骤一、谱剩余算法计算：Step 1. Spectral residual algorithm calculation:
首先对输入的灰度图像进行二维离散傅里叶变换，将图像从空间域转入频率域:，其中，为灰度图像空间域坐标，为灰度图像频率域坐标；First, two-dimensional discrete Fourier transform is performed on the input grayscale image, and the image is transferred from the spatial domain to the frequency domain: ,in, is the spatial domain coordinates of the grayscale image, is the grayscale image frequency domain coordinates;
(2) (2)
其中，为该点傅里叶频谱值。再求幅度谱和相位谱：in, is the Fourier spectrum value of this point. Then find the magnitude spectrum and phase spectrum:
(3) (3)
(4) (4)
对幅度谱取对数，得到其幅值的Log谱：Take the logarithm of the magnitude spectrum to get the Log spectrum of its magnitude:
(5) (5)
然后对Log谱进行平滑滤波，获取Log谱的大概形状：Then smooth and filter the Log spectrum to obtain the approximate shape of the Log spectrum:
(6) (6)
其中，是一个的平滑滤波器，为平滑滤波器的空域带宽。求取两者的差值，得到谱残差：in, Is a smoothing filter, is the spatial bandwidth of the smoothing filter. Find the difference between the two to get the spectral residual:
(7) (7)
对谱残差和相位谱进行二维傅里叶逆变换，得pair spectral residual and phase spectrum Carrying out two-dimensional inverse Fourier transform, we get
(8) (8)
其中，表示灰度图像中每点坐标的显著值。in, Indicates the saliency value of each point coordinate in the grayscale image.
步骤二、阈值的设定：Step 2. Threshold setting:
本发明采用两种方法设置阈值，并分别将阈值与同一幅图像中各像素点的显著值对比，将显著值大于等于阈值的像素点标记为“1”，记为目标区域；显著值小于阈值的像素点标记为“0”，记为背景区域；The present invention adopts two methods to set the threshold , and respectively compare the threshold with the salient value of each pixel in the same image, mark the pixel with a salient value greater than or equal to the threshold as "1" and record it as the target area; mark the pixel with a salient value smaller than the threshold as "0" , recorded as the background area;
根据自适应阈值算法求解。将阈值设为给定图像的平均显著值：Solve according to adaptive threshold algorithm . will threshold Set to mean saliency for a given image:
(9) (9)
其中，、对应图像的长和宽。将获取的每处显著值与自适应阈值比较，通过对比，将大于等于阈值的像素点标记为“1”，小于阈值的像素点标记为“0”， 并把该幅图像中的所有像素点集合用由0、1组成的行列矩阵表示。in, , corresponds to the height and width of the image. Combine each significant value obtained with an adaptive threshold Comparison, through comparison, mark the pixels greater than or equal to the threshold as "1", and mark the pixels less than the threshold as "0", and use the set of all pixels in the image to be composed of 0 and 1 Row column matrix express.
根据大律法求解。将图像显著值范围记为,(为最大显著值)。预先设置一阈值将上述图像显著值分为两类：,，并将与分别记为目标与背景，两者的类间方差为：Solve according to the great law . Denote the range of saliency values of the image as ,( is the maximum significant value). Preset a threshold The above image saliency values are divided into two categories: , , and will and They are respectively recorded as the target and the background, and the variance between the two classes is:
(10) (10)
(11) (11)
(12) (12)
其中，表示图像中显著值低于的像素个数，表示图像中显著值高于等于的像素个数，为低于的总像素的平均显著值，为高于等于的总像素的平均显著值；in, Indicates that the significant value in the image is lower than the number of pixels, Indicates that the significant value in the image is higher than or equal to the number of pixels, for less than The average saliency value of the total pixels of for higher than or equal to The average saliency value of the total pixels of ;
使得值最大的值就是所需的阈值，即。再将T与图像的每处显著值比较，将大于等于阈值的像素点标记为“1”，小于阈值的像素点标记为“0”， 并把该幅图像中的所有像素点集合用由0、1组成的行列矩阵表示。make the largest value value is the desired threshold ,Right now . Then compare T with each significant value of the image, mark the pixels greater than or equal to the threshold as "1", and mark the pixels less than the threshold as "0", and use all the pixel points in the image to be set by 0 , composed of 1 Row column matrix express.
步骤三、标记缺陷Step 3. Mark defects
将矩阵与矩阵对应元素相乘，得新矩阵。即：。由矩阵点乘公式可知，矩阵也是由元素0、1构成，元素1表示该像素点显著值均大于等于两阈值，即目标图像的重合位置。再按照从左往右、从上往下的顺序依次寻找三个矩阵、、中元素为1的连通区，分别将每个矩阵的连通区记为、、 。选取矩阵，对矩阵中所有元素求和，记和为；对矩阵中所有元素求和，记和为。引入函数，令 (13)the matrix with matrix Multiply the corresponding elements to get a new matrix . which is: . According to the matrix dot product formula, the matrix It is also composed of elements 0 and 1, and element 1 indicates that the salient values of the pixel are greater than or equal to the two thresholds , which is the overlapping position of the target image. Then search for three matrices in order from left to right and from top to bottom , , In the connected area where the element is 1, the connected area of each matrix is recorded as , , . selection matrix , for the matrix The sum of all elements in the sum is recorded as ; pair matrix The sum of all elements in the sum is recorded as . import function ,make (13)
若，则认为在r处为缺陷位置。若小于，则认为是误差检测，工件此处无缺陷。like , then it is considered that the position at r is the defect position. if less than , it is considered as an error detection, and the workpiece has no defects here.
本发明中，图像预处理、图像边缘检测、图像分割技术属于本领域类公知常识，本领域内的技术人员可根据被测物的具体要求再现，在此不再赘述。In the present invention, image preprocessing, image edge detection, and image segmentation technologies belong to the common knowledge in the field, and those skilled in the art can reproduce them according to the specific requirements of the measured object, and will not repeat them here.
尽管上面结合图对本发明进行了描述，但是本发明并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是局限性的，本领域的普通技术人员在本发明的启示下，在不脱离本发明宗旨的情况下，还可以作出很多变形，这些均属于本发明的保护之内。Although the present invention has been described above in conjunction with the drawings, the present invention is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative and not limiting. Under the inspiration, many modifications can be made without departing from the gist of the present invention, and these all belong to the protection of the present invention.
Claims (5)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611007125.2A CN106442556A (en) | 2016-11-16 | 2016-11-16 | Device and method for detecting surface defects of perforated plate workpiece |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611007125.2A CN106442556A (en) | 2016-11-16 | 2016-11-16 | Device and method for detecting surface defects of perforated plate workpiece |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106442556A true CN106442556A (en) | 2017-02-22 |
Family
ID=58207868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611007125.2A Pending CN106442556A (en) | 2016-11-16 | 2016-11-16 | Device and method for detecting surface defects of perforated plate workpiece |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106442556A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107014819A (en) * | 2017-06-09 | 2017-08-04 | 杭州电子科技大学 | A kind of solar panel surface defects detection system and method |
| CN107680086A (en) * | 2017-09-27 | 2018-02-09 | 电子科技大学 | A kind of existing arc-shaped side has the material profile defect inspection method of straight line again |
| CN108732186A (en) * | 2018-07-20 | 2018-11-02 | 梧州学院 | Embedded Surface Flaw automatic checkout system and its control method |
| CN109447989A (en) * | 2019-01-08 | 2019-03-08 | 哈尔滨理工大学 | Defect detecting device and method based on motor copper bar burr growth district |
| CN109584239A (en) * | 2018-12-13 | 2019-04-05 | 华南理工大学 | A kind of bloom body surface defect detecting system and method based on reflected light |
| CN109709105A (en) * | 2019-01-18 | 2019-05-03 | 安徽工程大学 | A kind of hole missing detection device |
| CN110288561A (en) * | 2018-03-14 | 2019-09-27 | 浙江大学山东工业技术研究院 | Scratch recognition method on refractory brick surface based on frequency domain filter enhancement |
| CN111272766A (en) * | 2020-02-20 | 2020-06-12 | 上海普密德自动化科技有限公司 | Surface defect detection system based on vision technology and detection method thereof |
| CN111487192A (en) * | 2020-04-26 | 2020-08-04 | 天津海融科技有限公司 | Machine vision surface defect detection device and method based on artificial intelligence |
| CN112859189A (en) * | 2020-12-31 | 2021-05-28 | 广东美的白色家电技术创新中心有限公司 | Workpiece detection device, detection method, and computer-readable storage medium |
| CN113538432A (en) * | 2021-09-17 | 2021-10-22 | 南通蓝城机械科技有限公司 | Part defect detection method and system based on image processing |
| CN113592787A (en) * | 2021-07-13 | 2021-11-02 | 苏州汇川控制技术有限公司 | Light emitting component detection method, light emitting component detection device, terminal equipment and storage medium |
| CN114663430A (en) * | 2022-05-18 | 2022-06-24 | 爱科赛智能科技(浙江)有限公司 | PCB surface defect detection method based on frequency domain information double confirmation |
| CN114964007A (en) * | 2022-06-03 | 2022-08-30 | 哈尔滨理工大学 | Visual measurement and surface defect detection method for weld size |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005172559A (en) * | 2003-12-10 | 2005-06-30 | Seiko Epson Corp | Line defect detection method and apparatus for panel |
| CN102305793A (en) * | 2011-05-11 | 2012-01-04 | 苏州天准精密技术有限公司 | Method and equipment for detecting appearance quality of product |
| CN102495069A (en) * | 2011-12-07 | 2012-06-13 | 广东辉丰科技股份有限公司 | Method for detecting defects of chain belts of zipper on basis of digital image processing |
| CN102590330A (en) * | 2011-12-29 | 2012-07-18 | 南京理工大学常熟研究院有限公司 | Intelligent identification and detection system for magnetic particle flaw detection defects based on image processing |
| CN104360501A (en) * | 2014-10-15 | 2015-02-18 | 西安交通大学 | Visual detection method and device for defects of liquid crystal display screen |
| CN105181713A (en) * | 2015-07-19 | 2015-12-23 | 中北大学 | Detection device used for optical fiber image inverter surface defects |
| CN106053479A (en) * | 2016-07-21 | 2016-10-26 | 湘潭大学 | System for visually detecting workpiece appearance defects based on image processing |
| CN206223683U (en) * | 2016-11-16 | 2017-06-06 | 哈尔滨理工大学 | A kind of tabular workpiece with hole surface defect detection apparatus |
-
2016
- 2016-11-16 CN CN201611007125.2A patent/CN106442556A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005172559A (en) * | 2003-12-10 | 2005-06-30 | Seiko Epson Corp | Line defect detection method and apparatus for panel |
| CN102305793A (en) * | 2011-05-11 | 2012-01-04 | 苏州天准精密技术有限公司 | Method and equipment for detecting appearance quality of product |
| CN102495069A (en) * | 2011-12-07 | 2012-06-13 | 广东辉丰科技股份有限公司 | Method for detecting defects of chain belts of zipper on basis of digital image processing |
| CN102590330A (en) * | 2011-12-29 | 2012-07-18 | 南京理工大学常熟研究院有限公司 | Intelligent identification and detection system for magnetic particle flaw detection defects based on image processing |
| CN104360501A (en) * | 2014-10-15 | 2015-02-18 | 西安交通大学 | Visual detection method and device for defects of liquid crystal display screen |
| CN105181713A (en) * | 2015-07-19 | 2015-12-23 | 中北大学 | Detection device used for optical fiber image inverter surface defects |
| CN106053479A (en) * | 2016-07-21 | 2016-10-26 | 湘潭大学 | System for visually detecting workpiece appearance defects based on image processing |
| CN206223683U (en) * | 2016-11-16 | 2017-06-06 | 哈尔滨理工大学 | A kind of tabular workpiece with hole surface defect detection apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 陈向伟;张学军;关山;: "基于计算机视觉的微小轴承表面缺陷检测", 机床与液压 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107014819A (en) * | 2017-06-09 | 2017-08-04 | 杭州电子科技大学 | A kind of solar panel surface defects detection system and method |
| CN107680086A (en) * | 2017-09-27 | 2018-02-09 | 电子科技大学 | A kind of existing arc-shaped side has the material profile defect inspection method of straight line again |
| CN107680086B (en) * | 2017-09-27 | 2020-10-23 | 电子科技大学 | Method for detecting material contour defects with arc-shaped edges and linear edges |
| CN110288561A (en) * | 2018-03-14 | 2019-09-27 | 浙江大学山东工业技术研究院 | Scratch recognition method on refractory brick surface based on frequency domain filter enhancement |
| CN108732186A (en) * | 2018-07-20 | 2018-11-02 | 梧州学院 | Embedded Surface Flaw automatic checkout system and its control method |
| CN109584239A (en) * | 2018-12-13 | 2019-04-05 | 华南理工大学 | A kind of bloom body surface defect detecting system and method based on reflected light |
| CN109584239B (en) * | 2018-12-13 | 2024-02-06 | 华南理工大学 | High-light object surface defect detection system and method based on reflected light |
| CN109447989A (en) * | 2019-01-08 | 2019-03-08 | 哈尔滨理工大学 | Defect detecting device and method based on motor copper bar burr growth district |
| CN109709105B (en) * | 2019-01-18 | 2021-07-27 | 安徽工程大学 | A hole missing detection device |
| CN109709105A (en) * | 2019-01-18 | 2019-05-03 | 安徽工程大学 | A kind of hole missing detection device |
| CN111272766A (en) * | 2020-02-20 | 2020-06-12 | 上海普密德自动化科技有限公司 | Surface defect detection system based on vision technology and detection method thereof |
| CN111487192A (en) * | 2020-04-26 | 2020-08-04 | 天津海融科技有限公司 | Machine vision surface defect detection device and method based on artificial intelligence |
| CN112859189A (en) * | 2020-12-31 | 2021-05-28 | 广东美的白色家电技术创新中心有限公司 | Workpiece detection device, detection method, and computer-readable storage medium |
| CN113592787A (en) * | 2021-07-13 | 2021-11-02 | 苏州汇川控制技术有限公司 | Light emitting component detection method, light emitting component detection device, terminal equipment and storage medium |
| CN113538432A (en) * | 2021-09-17 | 2021-10-22 | 南通蓝城机械科技有限公司 | Part defect detection method and system based on image processing |
| CN113538432B (en) * | 2021-09-17 | 2021-12-21 | 南通蓝城机械科技有限公司 | Part defect detection method and system based on image processing |
| CN114663430A (en) * | 2022-05-18 | 2022-06-24 | 爱科赛智能科技(浙江)有限公司 | PCB surface defect detection method based on frequency domain information double confirmation |
| CN114964007A (en) * | 2022-06-03 | 2022-08-30 | 哈尔滨理工大学 | Visual measurement and surface defect detection method for weld size |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106442556A (en) | Device and method for detecting surface defects of perforated plate workpiece | |
| CN102176228B (en) | A Machine Vision Method for Recognizing Information on Multiple Analog Instrument Dials | |
| CN109816644B (en) | An automatic detection system for bearing defects based on multi-angle light source images | |
| CN105067638B (en) | Tire fetal membrane face character defect inspection method based on machine vision | |
| CN107941808B (en) | 3D printing forming quality detection system and method based on machine vision | |
| CN110403232B (en) | Cigarette quality detection method based on secondary algorithm | |
| CN110108711A (en) | Visual inspection system for defects of side wall of circular ring | |
| CN111353993B (en) | A thread angle measurement method based on machine vision | |
| CN103454285A (en) | Transmission chain quality detection system based on machine vision | |
| CN116416250B (en) | Finished product detecting system of fast food canned product production line | |
| CN109507192A (en) | A kind of magnetic core detection method of surface flaw based on machine vision | |
| CN103839283A (en) | Area and circumference nondestructive measurement method of small irregular object | |
| CN113252568A (en) | Lens surface defect detection method, system, product and terminal based on machine vision | |
| CN108802052A (en) | A kind of detecting system and its detection method about slide fastener defect | |
| CN105095958A (en) | Cocoon counting method | |
| CN111179362B (en) | Test paper color uniformity detection method based on dynamic illumination correction algorithm | |
| CN107240088A (en) | Detection dividing method, system and the device of the adhesion grain of rice | |
| CN117333489B (en) | Film damage detection device and detection system | |
| WO2024016632A1 (en) | Bright spot location method, bright spot location apparatus, electronic device and storage medium | |
| KR102030768B1 (en) | Poultry weight measuring method using image, recording medium and device for performing the method | |
| CN108416814A (en) | A method and system for fast positioning and identification of pineapple heads | |
| CN206223683U (en) | A kind of tabular workpiece with hole surface defect detection apparatus | |
| CN111562260B (en) | A method and device for detecting lotus root mud holes based on machine vision | |
| TW201601119A (en) | Method for recognizing and locating object | |
| CN110084587B (en) | Automatic dinner plate settlement method based on edge context |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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: 20170222 |