WO2017067342A1 - 板卡位置检测方法及装置 - Google Patents
板卡位置检测方法及装置 Download PDFInfo
- Publication number
- WO2017067342A1 WO2017067342A1 PCT/CN2016/098222 CN2016098222W WO2017067342A1 WO 2017067342 A1 WO2017067342 A1 WO 2017067342A1 CN 2016098222 W CN2016098222 W CN 2016098222W WO 2017067342 A1 WO2017067342 A1 WO 2017067342A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- contour
- area
- maximum value
- edge
- Prior art date
Links
- 238000001514 detection method Methods 0.000 title abstract description 27
- 230000000877 morphologic effect Effects 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000003708 edge detection Methods 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims description 38
- 238000004364 calculation method Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 230000010339 dilation Effects 0.000 description 6
- 230000003628 erosive effect Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 239000007787 solid Substances 0.000 description 3
- 230000000295 complement effect Effects 0.000 description 2
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 238000007689 inspection 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
- 229910000679 solder Inorganic materials 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
Definitions
- the invention relates to the field of automatic optical detection, and in particular to a method and a device for detecting a position of a card.
- AOI Automatic Optic Inspection
- PCB Print circuit board
- the AOI system moves the PCB card to the shooting range of the camera through the pipeline conveyor.
- the light of the light source is irradiated on the PCB and reflected into the camera.
- the camera obtains an image reflecting the characteristics of the component according to the intensity of the reflected light, and then passes through
- the computer uses image processing technology to compare the images and obtain the detection results.
- the AOI system needs to detect whether the PCB board has moved into the shooting range of the camera, and automatically intercepts the board picture to initiate subsequent processes such as detecting components.
- the AOI system generally uses external sensors such as infrared and laser to detect the position of the board, but in the implementation process, the inventors found that at least the following problems exist in the conventional technology:
- the AOI system uses an external sensor to determine whether the board enters the camera's shooting range, which increases the complexity and cost of the system and does not capture the exact position of the board.
- an embodiment of the technical solution of the present invention is:
- a method for detecting a position of a card including the following steps:
- the center point coordinates of the connected area corresponding to the maximum value are obtained, and the center point coordinates are used as the center point coordinates of the board.
- a card position detecting device including:
- An image acquisition module configured to capture an image of the location to be detected, and obtain a grayscale image of the image
- An image processing module configured to perform edge detection processing on the grayscale image to obtain an edge map of the grayscale image; and obtain a morphological image of the edge map including a plurality of connected regions; and calculate an area of each connected region in the morphological image;
- a determining module configured to obtain a maximum value in the area, and determine whether the maximum value is greater than or equal to a preset area threshold
- the position obtaining module is configured to obtain a center point coordinate of the connected area corresponding to the maximum value when the judgment result of the determining module is YES, and use the center point coordinate as the center point coordinate of the board.
- the board position detection method and device of the invention enable the AOI system to determine whether there is a card in the picture taken by the camera, thereby determining whether the board enters the camera shooting range; and can capture the board in the picture by using the obtained positioning information.
- the image of the card acquires the precise position of the board, and then sends the image into the detection algorithm to realize the detection of the PCB component; thereby enabling the AOI system to judge the position of the board by software method, and realize the entry detection of the board on the pipeline. Accurate positioning of the board reduces costs.
- Embodiment 1 is a flowchart of Embodiment 1 of a board position detecting method according to the present invention
- FIG. 2 is a flow chart of a specific embodiment of a board position detecting method according to the present invention.
- FIG. 3 is an image of a grayscale image acquired in a specific embodiment of a board position detecting method according to the present invention.
- FIG. 4 is an image of an edge map in a specific embodiment of a board position detecting method according to the present invention.
- FIG. 5 is a morphological image of a specific embodiment of a board position detecting method according to the present invention.
- FIG. 6 is a schematic structural view of Embodiment 1 of a board position detecting device according to the present invention.
- FIG. 1 is a flowchart of Embodiment 1 of the board position detecting method of the present invention; as shown in FIG. The method includes the following steps:
- Step S110 capturing an image of the position to be detected, and acquiring a grayscale image of the image
- Step S120 Perform edge detection processing on the grayscale image to obtain an edge map of the grayscale image, obtain a morphological image of the edge map including a plurality of connected regions, and calculate an area of each connected region in the morphological image;
- the edge detection process can distinguish the effective object points (the debris on the PCB board or other pipelines) from the pipeline (background image) in the image acquired within the shooting range of the AOI system; this step Through the mathematical morphology operation of the edge map, the cavity in the middle of the edge map of the card can be filled, and the edge image of each effective object point can be connected into a morphological image having a plurality of white connected regions, so as to calculate the connected region. Area.
- Step S130 obtaining the maximum value in the area, and determining whether the maximum value is greater than or equal to the preset area threshold; if the determination result is yes, proceeding to step S140;
- Step S140 Acquire a center point coordinate of the connected area corresponding to the maximum value, and use the center point coordinate as the center point coordinate of the board.
- steps S130 to S140 are to screen the area with the area less than the value by using the preset area threshold. If the maximum value is smaller than the preset area threshold, it can be considered that there is no PCB card in the picture taken by the camera; specifically, if There is no board in the figure, only the background pattern of the solid color, then the white connected area cannot be obtained in step S120, or only some areas whose area is smaller than the preset area threshold. In the embodiment of the board position detecting method of the present invention, if there is no PCB card in the image acquired by the current camera, but some noise is generated in the step S120 due to noise or other debris, These areas will not be judged as boards.
- step S140 may include:
- the method for calculating the coordinates of the center point includes two types:
- the center point coordinate of the connected area is the precise position of the PCB board.
- the AOI system can judge whether the PCB board moves within the shooting range of the camera according to the precise position, and automatically intercepts the board picture, thereby starting the detection component and the like. process.
- the step of calculating the area of each connected region in the morphological image in step S120 includes:
- Each connected region in the morphological image is processed to obtain a contour corresponding to each connected region; an area is calculated for the contour to obtain an area of the contour.
- the area of each contour is the area of each connected area.
- the step of acquiring the morphological image of the edge map and including the connected regions in step S120 can be implemented by performing a mathematical morphology operation on the edge map, including:
- the edge map and the operation kernel are convoluted; in the embodiment of the present invention, the operation kernel selects a rectangular core; specifically, the edge map is subjected to morphological closed operation, and the operation kernel is A rectangular core of n ⁇ n or n ⁇ m, to obtain a morphological image; the shape of the computing kernel affects the result graph. For example, if the computing kernel is a circular core, the corners of the morphology-processed graph will be arcs. Because the PCB board is rectangular, the use of a rectangular core results in a more similar result to the original image.
- the board position detection method of the present invention enables the AOI system to determine whether there is a card in the picture taken by the camera, thereby determining whether the board enters the camera shooting range; and can intercept the board in the picture by using the obtained positioning information.
- the image acquires the precise position of the board, and then sends the image into the detection algorithm to realize the detection of the PCB component; thereby enabling the AOI system to judge the position of the board by software method, realize the entry detection of the board on the pipeline, and realize the board.
- the precise positioning of the card reduces costs.
- FIG. 2 is a specific method for detecting the position of the board of the present invention.
- A. Obtain the grayscale image in the shooting range of the camera on the assembly line by camera (or other shooting device), I(x, y), x ⁇ [0, w), y ⁇ [0, h), h is the grayscale image.
- the height of the image, w is the width of the grayscale image.
- the grayscale image is specifically shown in FIG. 3, and FIG. 3 is an image of the grayscale image acquired in a specific embodiment of the method for detecting the position of the card;
- FIG. 4 is a specific embodiment of the method for detecting the position of the card in the present invention.
- the image of the edge map obtained based on the Canny algorithm; and the Canny and Sobel edge detection algorithms are commonly used algorithms in graphics.
- FIG. 5 shows the morphological image in a particular embodiment the card position detecting method embodiment of the invention
- the morphological operation of the edge image I edge may be performed by using an expansion operation to obtain a morphological image I' edge ; the expansion operation connects the disconnected target object, but after the expansion process, the target object The area is larger than the original area, which will cause errors in the positioning of the PCB board. Therefore, in the embodiment of the present invention, the morphological closing operation is preferentially selected as the processing scheme for acquiring the morphological image.
- the specific function of the arithmetic kernel in the morphological operation is to convolve the target image A (in the embodiment of the present invention, the edge map I edge ) with the kernel B.
- Dilation(A, B) (Erosion(A c , B)) c , where A c represents the complement of A;
- Closing (A, B) Erosion (Dilation (A, B), B);
- the shape of the core B affects the result graph. For example, if the computing kernel is a circular core, the corners of the morphologically processed graph will be arcs. Because the PCB board is rectangular, using a rectangular core results in a more similar result to the original image.
- the rectangular core n ⁇ n may also be n ⁇ m.
- function cvFindContours in the OpenCV (open source computer vision) library for S i (OpenCV image processing library: function cvFindContours retrieves the contour from the binary image and returns the number of detected contours) to obtain S i The contour of C i . Then use the function contourArea (the function cvContourArea is used to calculate the area of the whole or part of the contour) to get the area of C i .
- Area threshold indicates the preset area threshold, which is used to screen out the area where the area is less than the value; the Area threshold is the empirical value of the experiment, and its size is between 0 and w*h.
- the general PCB board shape is a rectangle.
- the coordinate origin is in the upper left corner of the image, the origin is the x-axis direction in the right direction, and the origin is in the y-axis direction in the downward direction, so get:
- Step F and G is achieved by the following methods: using a library function cvFindContours OpenCV to obtain S i C i S i of the contour. Then use the function boundingRect (calculate the outermost (up-right) boundary of the point set) to get the maximum point coordinates of the corners of the circumscribed matrices of C i , and finally calculate the coordinates of the center point of the board according to the above formula; you can also use the function boundingRect Obtain the coordinates of the upper left corner of the circumscribed matrix of C i (x left-up , y left-up ), and the length w ⁇ and width h ⁇ .
- a library function cvFindContours OpenCV to obtain S i C i S i of the contour. Then use the function boundingRect (calculate the outermost (up-right) boundary of the point set) to get the maximum point coordinates of the corners of the circumscribed matrices of C i , and finally calculate
- other corner coordinates of the circumscribed matrix of C i , and the length w′ and the width h′ can also be obtained, and the coordinates of the center point are calculated by the corresponding formula.
- the board position detection method of the present invention enables the AOI system to determine whether there is a card in the picture taken by the camera, thereby determining whether the board enters the camera shooting range;
- the positioning information intercepts the image of the board in the picture, obtains the precise position of the board, and then sends the image into the detection algorithm to realize the detection of the PCB component; thereby enabling the AOI system to determine the position of the board by software method and realize the pipeline.
- the entry detection of the upper board enables precise positioning of the board and reduces costs.
- Embodiment 1 of the board position detecting device of the present invention is a diagrammatic representation of the board position detecting device of the present invention.
- FIG. 6 is a schematic structural view of Embodiment 1 of the board position detecting device of the present invention, as shown in FIG.
- the detection device includes:
- the image acquisition module 610 is configured to capture an image of the location to be detected, and obtain a grayscale image of the image; in a specific embodiment, obtain a grayscale image of an image captured by the pipeline camera of the AOI system;
- the image processing module 620 is configured to perform edge detection processing on the grayscale image to obtain an edge map of the grayscale image; and obtain a morphology image of the edge map including a plurality of connected regions; and calculate an area of each connected region in the morphology image
- the edge detection processing is performed on the grayscale image to obtain an edge map of the grayscale image
- the edge morphology is subjected to mathematical morphology processing to obtain a morphological image
- Each connected region performs a function operation process to obtain an area of each connected region;
- the determining module 630 is configured to obtain a maximum value in the area, and determine whether the maximum value is greater than or equal to a preset area threshold;
- the location obtaining module 640 is configured to obtain a maximum value when the determination result of the determining module 630 is YES.
- the coordinates of the center point of the connected area and the coordinates of the center point are taken as the coordinates of the center point of the board.
- the location obtaining module 640 includes:
- a first contour obtaining module 642 configured to acquire a contour of the connected area corresponding to the maximum value
- the location obtaining module 640 can also acquire the center point coordinates by using the first contour acquiring module 642 and the second coordinate acquiring module 645; the second coordinate acquiring module 645 is configured to calculate the corner coordinates and the contour of the contour. Length and width; and get the coordinates of the center point according to the corner coordinates and length and width.
- a second contour acquiring module 622 configured to acquire a contour of each connected region in the morphological image
- the area obtaining module 624 is configured to perform area calculation on the contour to obtain an area of the contour.
- the image processing module 620 uses the Canny or Sobel edge detection algorithm to obtain the edge map I edge of the grayscale image I(x, y); and the Canny and Sobel edge detection algorithms are commonly used algorithms in graphics.
- image processing module 620 of FIG. I edge for edge morphological closing process a rectangular core processor cores, to obtain morphology image I ⁇ edge; Specifically, i.e. diagrams edge morphological closing operation, processor cores I edge It is a rectangular core of n ⁇ n size. After this operation can be filled by the card edge view of the intermediate cavity, to obtain morphology image I ⁇ edge.
- the specific function of the arithmetic kernel in the morphological operation is to convolve the target image A (in the embodiment of the present invention, the edge map I edge ) with the kernel B.
- Dilation(A, B) (Erosion(A c , B)) c , where A c represents the complement of A;
- the shape of the core B affects the result graph. For example, if the computing kernel is a circular core, the corners of the morphologically processed graph will be arcs. Because the PCB board is rectangular, using a rectangular core results in a more similar result to the original image.
- the rectangular core n ⁇ n may also be n ⁇ m.
- Area max is smaller than Area threshold
- Area i , i ⁇ (0, 1, ..., m) is the number of pixels in each connected area, indicating the area of the area.
- the Area threshold indicates the preset area threshold, which is used to screen out the area where the area is less than the value; the Area threshold is the empirical value of the experiment, and its size is between 0 and w*h.
- the first coordinate acquiring module 644 in the position obtaining module 640 can obtain the highest point coordinates of the upper left, lower left, upper right, and lower right corners of S max as follows:
- the general PCB card shape is a rectangle.
- the coordinate origin is in the upper left corner of the image, the origin is the x-axis direction in the right direction, and the origin is in the y-axis direction in the downward direction. This can be obtained:
- the first coordinate obtaining module 644 obtains the maximum point coordinates of the corners of the circumscribed matrices of C i using the function boundingRect (calculating the outermost (up-right) rectangular boundary of the point set), and finally calculates the coordinates of the center point of the board according to the above formula. ;
- the second coordinate acquisition module 645 obtains the upper left coordinate (x left-up , y left-up ) of the circumscribed matrix of C i using the function boundingRect, and the length w ⁇ and the width h ′, and the second coordinate acquisition module 645
- the second coordinate acquisition module 645 can also obtain other corner coordinates of the circumscribed matrix of C i , and the length w′ and the width h′, and calculate the center point coordinates by a corresponding formula.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
一种板卡位置检测方法及装置。其中,板卡位置检测方法包括以下步骤:拍摄待检测位置的图像,获取该图像的灰度图(S110);对灰度图进行边缘检测处理,得到灰度图的边缘图,获取边缘图的包含若干连通区域的形态学图像,并计算形态学图像中的各连通区域的面积(S120);获取面积中的最大值,并判断最大值是否大于或等于预设面积阈值(S130);若判断结果为是,获取最大值对应的连通区域的中心点坐标,并将中心点坐标作为板卡的中心点坐标(S140)。所述板卡位置检测方法及装置使AOI系统能够通过软件方法判断板卡位置,实现流水线上板卡的进入检测,实现对板卡的精确定位,降低成本。
Description
本发明涉及自动光学检测领域,特别是涉及一种板卡位置检测方法及装置。
AOI(Automatic Optic Inspection,自动光学检测)设备是基于光学原理来对焊接生产中遇到的常见缺陷进行检测的设备。当自动检测时,机器通过摄像头自动扫描PCB(Printed circuit board,印制电路板),采集图像,测试的焊点与数据库中的合格的参数进行比较,经过图像处理,检查出PCB上的缺陷,并通过显示器或自动标志把缺陷显示/标示出来,供维修人员修整。
具体而言,AOI系统通过流水线传送带将PCB板卡移动到摄像头的拍摄范围内,光源的光照射在PCB板上并反射进入摄像机,摄像机根据反射光的强弱得到反映元件特征的图像,然后通过计算机采用图像处理技术对图像进行比对处理后得到检测结果。AOI系统需要探测PCB板卡是否移动到摄像头的拍摄范围内,并且自动截取板卡图片,从而启动检测元件等后续过程。AOI系统一般使用红外、激光等外部传感器探测板卡位置,但在实现过程中,发明人发现传统技术中至少存在如下问题:
AOI系统使用外部传感器判断板卡是否进入摄像头拍摄范围,会增加系统的复杂度和成本,且不能获取板卡的精确位置。
发明内容
基于此,有必要针对AOI系统检测板卡位置系统的复杂度与成本较高的问题,提供一种板卡位置检测方法及装置。
为了实现上述目的,本发明技术方案的实施例为:
一方面,提供了一种板卡位置检测方法,包括以下步骤:
拍摄待检测位置的图像,获取该图像的灰度图;
对灰度图进行边缘检测处理,得到灰度图的边缘图,获取边缘图的包含若
干连通区域的形态学图像,并计算形态学图像中的各连通区域的面积;
获取面积中的最大值,并判断最大值是否大于或等于预设面积阈值;
若判断结果为是,获取最大值对应的连通区域的中心点坐标,并将中心点坐标作为板卡的中心点坐标。
另一方面,提供了一种板卡位置检测装置,包括:
图像获取模块,用于拍摄待检测位置的图像,获取该图像的灰度图;
图像处理模块,用于对灰度图进行边缘检测处理,得到灰度图的边缘图;并获取边缘图的包含若干连通区域的形态学图像;以及计算形态学图像中的各连通区域的面积;
判断模块,用于获取面积中的最大值,并判断最大值是否大于或等于预设面积阈值;
位置获取模块,用于在判断模块的判断结果为是时,获取最大值对应的连通区域的中心点坐标,并将中心点坐标作为板卡的中心点坐标。
上述技术方案具有如下有益效果:
通过本发明的板卡位置检测方法及装置,使AOI系统能够判断摄像头拍摄的图片中是否有板卡,从而判断板卡是否进入摄像头拍摄范围;并且能够通过所获得的定位信息在图片中截取板卡的图像,获取板卡的精确位置,然后将图像送入检测算法中,实现对PCB元件的检测;进而使AOI系统能够通过软件方法判断板卡位置,实现流水线上板卡的进入检测,实现对板卡的精确定位,降低成本。
通过附图中所示的本发明的优选实施例的更具体说明,本发明的上述及其它目的、特征和优势将变得更加清晰。在全部附图中相同的附图标记指示相同的部分,且并未刻意按实际尺寸等比例缩放绘制附图,重点在于示出本发明的主旨。
图1为本发明板卡位置检测方法实施例1的流程图;
图2为本发明板卡位置检测方法一具体实施例的流程图;
图3为本发明板卡位置检测方法一具体实施例中获取的灰度图的图像;
图4为本发明板卡位置检测方法一具体实施例中边缘图的图像;
图5为本发明板卡位置检测方法一具体实施例中的形态学图像;
图6为本发明板卡位置检测装置实施例1的结构示意图。
为了便于理解本发明,下面将参照相关附图对本发明进行更全面的描述。附图中给出了本发明的首选实施例。但是,本发明可以以许多不同的形式来实现,并不限于本文所描述的实施例。相反地,提供这些实施例的目的是使对本发明的公开内容更加透彻全面。
需要说明的是,当一个元件被认为是“连接”另一个元件,它可以是直接连接到另一个元件并与之结合为一体,或者可能同时存在居中元件。本文所使用的术语“边缘检测”、“数学形态学运算”、“最大值”以及类似的表述只是为了说明的目的。
除非另有定义,本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的,不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。
板卡位置检测方法实施例1:
针对AOI系统判断板卡位置的技术问题,本发明提供了一种板卡位置检测方法,图1为本发明板卡位置检测方法实施例1的流程图;如图1所示,板卡位置检测方法包括以下步骤:
步骤S110:拍摄待检测位置的图像,获取该图像的灰度图;
步骤S120:对灰度图进行边缘检测处理,得到灰度图的边缘图,获取边缘图的包含若干连通区域的形态学图像,并计算形态学图像中的各连通区域的面积;此步骤中的边缘检测处理可以将AOI系统拍摄范围内获取的图像中有效物点(PCB板卡或其它流水线上的杂物)与流水线(背景图像)相区分;此步骤
中通过对该边缘图进行数学形态学运算能够填满板卡边缘图中间的空洞,可以将各有效物点的边缘图像连通成为拥有若干个白色的连通区域的形态学图像,以便于计算连通区域的面积。
步骤S130:获取面积中的最大值,并判断最大值是否大于或等于预设面积阈值;若判定结果为是,则进入步骤S140;
步骤S140:获取最大值对应的连通区域的中心点坐标,并将该中心点坐标作为板卡的中心点坐标。
而步骤S130至S140的目的在于,用预设面积阈值筛去面积少于该值的区域,若最大值小于预设面积阈值,可以认为摄像头拍摄的图中没有PCB板卡;具体而言,如果图中没有板卡,只有纯色的背景图案,那步骤S120中就不能获得白色的连通区域,或者只有一些面积小于预设面积阈值的区域。若当前摄像头获取的图像中没有PCB板卡,但是由于噪音或者有其他杂物,而导致在步骤S120中得到了一些面积较小的连通区域时,在本发明板卡位置检测方法的实施例中不会将这些区域判断为板卡。
在一个具体实施例中,步骤S140中获取所述最大值对应的连通区域的中心点坐标的步骤可以包括:
对最大值对应的连通区域进行边缘检测处理,得到最大值对应的连通区域的轮廓;
而为了得到该最大值对应的连通区域的轮廓的中心点坐标,在一个具体实施例中,计算中心点坐标的方法包括两种:
1)获取该轮廓的上、下、左、右的最值点坐标,将上述最值点坐标作为所述轮廓的边缘点坐标;对轮廓进行位置计算,得到轮廓的边缘点坐标;根据边缘点坐标,获取连通区域的中心点坐标。
2)对该轮廓进行位置计算,得到该轮廓的边角坐标(左上角坐标)和轮廓的长度与宽度;根据边角坐标和长度与宽度,获取中心点坐标。
而连通区域的中心点坐标即为PCB板卡的精确位置,AOI系统可根据该精确位置,判断PCB板卡是否移动到摄像头的拍摄范围内,并且自动截取板卡图片,从而启动检测元件等后续过程。
而步骤S120中计算形态学图像中的各连通区域的面积的步骤包括:
对形态学图像中各连通区域进行处理,得到各连通区域对应的轮廓;对轮廓进行面积计算,得到轮廓的面积。而各轮廓的面积即为各连通区域的面积。
步骤S120中获取边缘图的包含若干连通区域的形态学图像的步骤可以通过对边缘图进行数学形态学运算处理得以实现,具体包括:
基于数学形态学运算处理过程,将边缘图与运算内核进行卷积运算;在本发明实施例中,运算内核选用矩形核;具体而言,即对边缘图进行形态学闭运算处理,运算内核为n×n或n×m大小的矩形核,从而得到形态学图像;运算内核的形状会影响结果图,例如若运算内核为圆形核,经过形态学处理的图的边角会是圆弧。因为PCB板卡是矩形的,所以使用矩形核能得到与原图像更相近的结果。
通过本发明的板卡位置检测方法,使AOI系统能够判断摄像头拍摄的图片中是否有板卡,从而判断板卡是否进入摄像头拍摄范围;并且能够通过所获得的定位信息在图片中截取板卡的图像,获取板卡的精确位置,然后将图像送入检测算法中,实现对PCB元件的检测;进而使AOI系统能够通过软件方法判断板卡位置,实现流水线上板卡的进入检测,实现对板卡的精确定位,降低成本。
本发明板卡位置检测方法一具体实施例:
为了更清楚的说明本发明的技术方案,提供了板卡位置检测方法一具体实施例来说明本发明如何解决AOI系统判断板卡位置的技术问题;图2为本发明板卡位置检测方法一具体实施例的流程图;如图2所示:
A、通过摄像机(或其他拍摄设备)获取流水线上摄像头拍摄范围内的灰度图,I(x,y),x∈[0,w),y∈[0,h),h为灰度图图像的高,w为灰度图图像的宽。灰度图图像具体如图3所示,图3为本发明板卡位置检测方法一具体实施例中获取的灰度图的图像;
B、使用Canny或Sobel边缘检测算法,得到灰度图I(x,y)的边缘图Iedge;边缘图图像具体如图4所示,图4为本发明板卡位置检测方法一具体实施例中基
于Canny算法得到的边缘图的图像;而Canny和Sobel缘检测算法在图形学中都是常用的算法。
C、对边缘图Iedge进行形态学闭运算处理,运算内核为矩形核,得到形态学图像I`edge;具体而言,即对边缘图Iedge进行形态学闭运算,运算内核为n×n大小的矩形核。通过该运算后能够填满板卡边缘图中间的空洞,获得形态学图像I`edge。I`edge的图像具体如图5所示,图5为本发明板卡位置检测方法一具体实施例中的形态学图像;
在一个具体实施例中,也可以利用膨胀运算对边缘图Iedge进行形态学运算处理,得到形态学图像I`edge;膨胀运算将断开的目标物进行接续,但经过膨胀处理之后,目标物的面积大于原有面积,会对PCB板卡定位造成误差。因此在本发明的实施例中优先选择形态学闭运算作为获取形态学图像的处理方案。
运算内核在形态学运算中的具体作用是将目标图像A(在本发明实施例中为边缘图Iedge)与核B进行卷积。
在膨胀运算中:Dilation(A,B)=(Erosion(Ac,B))c,其中,Ac表示A的补集;
在形态学闭运算中:Closing(A,B)=Erosion(Dilation(A,B),B);
而核心B的形状会影响结果图,例如若运算内核为圆形核,经过形态学处理的图的边角会是圆弧。因为PCB板卡是矩形的,所以使用矩形核心能得到与原图像更相近的结果。其中矩形核心n×n也可以为n×m。
D、计算形态学图像I`edge中连通区域Si(在一个具体的实施例中,如图5所
示,连通区域Si是白色的)的面积Areai,i∈(0,1,...,m),其中m为连通区域的数目。Areamax=max(Areai)为面积最大区域Smax,AOI系统认为该区域是PCB板卡所在的区域。
对Si使用OpenCV(open source computer vision:开源计算机视觉库)库中的函数cvFindContours(OpenCV图像处理库:函数cvFindContours从二值图像中检索轮廓,并返回检测到的轮廓的个数)获得Si的轮廓Ci。然后再使用函数contourArea(函数cvContourArea用于计算整个或部分轮廓的面积)获得Ci的面积。
E、判断Areamax是否小于Areathreshold,其中Areai,i∈(0,1,...,m)是每一个连通区域中像素点的个数,表示区域的面积,而在一个具体的实施例中,如图5所示,上述像素点为白色的。而Areathreshold表示预设面积阈值,用来筛去面积少于该值的区域;Areathreshold作为实验的经验值,其大小在0到w*h之间。
若Areamax小于Areathreshold,可以认为摄像头拍摄的图中没有PCB板卡;具体而言,如果图中没有板卡,只有纯色的背景图案,那步骤D中就不能获得连通区域,或者只有一些面积小于Areathreshold的区域。步骤E具体用于:若当前摄像头获取的图像中没有PCB板卡,但是由于噪音或者有其他杂物,而导致在步骤D中得到了一些面积较小的连通区域时,在本发明板卡位置检测方法的实施例中AOI系统不会将这些区域判断为板卡。
F、若Areamax大于或等于Areathreshold,获取Smax的左上、左下、右上、右下边角的最值点坐标,如下所示:
Pleft-up=(xleft-up,yleft-up);
Pleft-down=(xleft-down,yleft-down);
Pright-up=(xright-up,yright-up);
Pright-down=(xright-down,yright-down);
其中,一般PCB板卡形状为矩形,在本发明一实施例的图像坐标系中,坐标原点在图像的左上角,原点往右方向为x轴方向,原点往下方向为y轴方向,因此可以得到:
xleft-up=xleft-down;xright-up=xright-down;yleft-up=yright-up;yleft-down=yright-down;
G、通过上述4个最值点,获得板卡的中心点坐标Pcenter=(xcenter,ycenter),即为PCB板卡的位置:
具体而言,步骤F和G通过以下方法实现:对Si使用OpenCV库中的函数cvFindContours获得Si的轮廓Ci。然后再使用函数boundingRect(计算点集的最外面(up-right)矩形边界)获得Ci的外接矩阵边角的最值点坐标,最后根据上述公式计算板卡中心点坐标;也可以使用函数boundingRect获得Ci的外接矩阵
的左上角坐标(xleft-up,yleft-up),以及长w`和宽h`,此时步骤G中的中心点坐标为Pcenter=(xleft-up+w`/2,yleft-up+h`/2)。在一个具体的实施例中,也可以获取Ci的外接矩阵的其它边角坐标,以及以及长w`和宽h`,并通过相应的公式计算中心点坐标。
通过以上具体的实施例,可以看出通过本发明的板卡位置检测方法,使AOI系统能够判断摄像头拍摄的图片中是否有板卡,从而判断板卡是否进入摄像头拍摄范围;并且能够通过所获得的定位信息在图片中截取板卡的图像,获取板卡的精确位置,然后将图像送入检测算法中,实现对PCB元件的检测;进而使AOI系统能够通过软件方法判断板卡位置,实现流水线上板卡的进入检测,实现对板卡的精确定位,降低成本。
本发明板卡位置检测装置实施例1:
针对AOI系统判断板卡位置的技术问题,本发明还提供了一种板卡位置检测装置;图6为本发明板卡位置检测装置实施例1的结构示意图,如图6所示:板卡位置检测装置包括:
图像获取模块610,用于拍摄待检测位置的图像,获取该图像的灰度图;在一个具体的实施例中,即获取AOI系统流水线摄像头拍摄的图像的灰度图;
图像处理模块620,用于对灰度图进行边缘检测处理,得到灰度图的边缘图;并获取边缘图的包含若干连通区域的形态学图像;以及计算形态学图像中的各连通区域的面积;在一个具体的实施例中,即对灰度图进行边缘检测处理,得到灰度图的边缘图;并对边缘图进行数学形态学运算处理,得到形态学图像;以及对形态学图像中的各连通区域进行函数运算处理,得到各连通区域的面积;
判断模块630,用于获取面积中的最大值,并判断最大值是否大于或等于预设面积阈值;
位置获取模块640,用于在判断模块630的判断结果为是时,获取最大值对
应的连通区域的中心点坐标,并将中心点坐标作为板卡的中心点坐标。
其中,位置获取模块640包括:
第一轮廓获取模块642,用于获取最大值对应的连通区域的轮廓;
第一坐标获取模块644,用于计算上述轮廓的边缘点坐标;并根据边缘点坐标,获取中心点坐标。
在一个具体的实施例中,位置获取模块640还可以通过第一轮廓获取模块642以及第二坐标获取模块645获取中心点坐标;第二坐标获取模块645用于计算轮廓的边角坐标和轮廓的长度与宽度;并根据边角坐标和长度与宽度,获取中心点坐标。
而图像处理模块620包括:
第二轮廓获取模块622,用于获取形态学图像中各连通区域的轮廓;
面积获取模块624,用于对上述轮廓进行面积计算,得到轮廓的面积。
为了进一步说明本发明板卡位置检测装置的具体作用过程,特提供一具体实施例来具体说明:
A、图像获取模块610通过摄像机(或其他拍摄设备)获取流水线上摄像头拍摄范围内拍摄的图像的灰度图,I(x,y),x∈[0,w),y∈[0,h),h为灰度图图像的高,w为灰度图图像的宽。
B、图像处理模块620使用Canny或Sobel边缘检测算法,获取灰度图I(x,y)的边缘图Iedge;而Canny和Sobel缘检测算法在图形学中都是常用的算法。
C、图像处理模块620对边缘图Iedge进行形态学闭运算处理,运算内核为矩形核,得到形态学图像I`edge;具体而言,即对边缘图Iedge进行形态学闭运算,运算内核为n×n大小的矩形核。通过该运算后能够填满板卡边缘图中间的空洞,获得形态学图像I`edge。
在一个具体实施例中,图像处理模块620运用膨胀运算也可以对边缘图Iedge进行形态学运算处理,得到形态学图像I`edge;膨胀运算将断开的目标物进行接续,但经过膨胀处理之后,目标物的面积大于原有面积,会对PCB板卡定位造成误差。因此在本发明的实施例中优先选择形态学闭运算作为获取形态学图像的处理方案。
运算内核在形态学运算中的具体作用是将目标图像A(在本发明实施例中为边缘图Iedge)与核B进行卷积。
在膨胀运算中:Dilation(A,B)=(Erosion(Ac,B))c,其中,Ac表示A的补集;
在形态学闭运算中:Closing(A,B)=Erosion(Dilation(A,B),B);
而核心B的形状会影响结果图,例如若运算内核为圆形核,经过形态学处理的图的边角会是圆弧。因为PCB板卡是矩形的,所以使用矩形核心能得到与原图像更相近的结果。其中矩形核心n×n也可以为n×m。
D、图像处理模块620计算形态学图像I`edge中连通区域Si的面积Areai,i∈(0,1,...,m),其中m为连通区域的数目。Areamax=max(Areai)为面积最大区域Smax,AOI系统认为该区域是PCB板卡所在的区域。
第二轮廓获取模块622对Si使用OpenCV(open source computer vision:开源计算机视觉库)库中的函数cvFindContours(OpenCV图像处理库:函数cvFindContours从二值图像中检索轮廓,并返回检测到的轮廓的个数)获得Si的轮廓Ci。然后由面积获取模块624使用函数contourArea(函数cvContourArea用于计算整个或部分轮廓的面积)获得Ci的面积,判断模块630根据Ci的面积,获取该面积中的最大值Areamax。
E、若Areamax是否小于Areathreshold,其中Areai,i∈(0,1,...,m)是每一个连通区域中像素点的个数,表示区域的面积。而Areathreshold表示预设面积阈值,用来筛去面积少于该值的区域;Areathreshold作为实验的经验值,其大小在0到w*h之间。
若Areamax小于Areathreshold,可以认为摄像头拍摄的图中没有PCB板卡;具体而言,如果图中没有板卡,只有纯色的背景图案,那形态学运算模块624就不能通过形态学运算获得连通区域,或者只有一些面积小于Areathreshold的区域。若当前摄像头获取的图像中没有PCB板卡,但是由于噪音或者有其他杂物,而导致在形态学运算中得到了一些面积较小的连通区域时,在本发明板卡位置检测装置的实施例中AOI系统不会将这些区域判断为板卡。
F、若Areamax大于或等于Areathreshold,位置获取模块640中的第一坐标获取模块644可获取Smax的左上、左下、右上、右下边角的最值点坐标,如下所示:
Pleft-up=(xleft-up,yleft-up);
Pleft-down=(xleft-down,yleft-down);
Pright-up=(xright-up,yright-up);
Pright-down=(xright-down,yright-down);
其中,一般PCB板卡形状为矩形,在本发明一实施例的图像坐标系中,坐标原点在图像的左上角,原点往右方向为x轴方向,原点往下方向为y轴方向,因
此可以得到:
xleft-up=xleft-down;xright-up=xright-down;yleft-up=yright-up;yleft-down=yright-down;
G、通过上述4个最值点,第一坐标获取模块644获得板卡的中心点坐标Pcenter=(xcenter,ycenter),即为PCB板卡的位置:
具体而言,第一轮廓获取模块642对Si使用OpenCV库中的函数cvFindContours获得Si的轮廓Ci。
然后由第一坐标获取模块644使用函数boundingRect(计算点集的最外面(up-right)矩形边界)获得Ci的外接矩阵边角的最值点坐标,最后根据上述公式计算板卡中心点坐标;
或者由第二坐标获取模块645使用函数boundingRect获得Ci的外接矩阵的左上角坐标(xleft-up,yleft-up),以及长w`和宽h`,此时第二坐标获取模块645获取的中心点坐标为Pcenter=(xleft-up+w`/2,yleft-up+h`/2)。在一个具体的实施例中,第二坐标获取模块645也可以获取Ci的外接矩阵的其它边角坐标,以及以及长w`和宽h`,并通过相应的公式计算中心点坐标。
通过以上具体的实施例,可以看出通过本发明的板卡位置检测装置使AOI系统能够判断摄像头拍摄的图片中是否有板卡,从而判断板卡是否进入摄像头拍摄范围;并且能够通过所获得的定位信息在图片中截取板卡的图像,获取板卡的精确位置,然后将图像送入检测算法中,实现对PCB元件的检测;进而使AOI系统能够通过软件方法判断板卡位置,实现流水线上板卡的进入检测,实现对板卡的精确定位,降低成本。
以上所述实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本发明构思的前提下,还可以做出若干变形和改进,这些都属于本发明的保护范围。因此,本发明专利的保护范围应以所附权利要求为准。
Claims (10)
- 一种板卡位置检测方法,其特征在于,包括以下步骤:拍摄待检测位置的图像,获取所述图像的灰度图;对所述灰度图进行边缘检测处理,得到所述灰度图的边缘图,获取所述边缘图的包含若干连通区域的形态学图像,并计算所述形态学图像中的各连通区域的面积;获取所述面积中的最大值,并判断所述最大值是否大于或等于预设面积阈值;若所述判断结果为是,获取所述最大值对应的连通区域的中心点坐标,并将所述中心点坐标作为所述板卡的中心点坐标。
- 根据权利要求1所述的板卡位置检测方法,其特征在于,所述获取所述最大值对应的连通区域的中心点坐标的步骤包括:获取所述最大值对应的连通区域的轮廓;计算所述轮廓的边缘点坐标;根据所述边缘点坐标,获取所述中心点坐标。
- 根据权利要求2所述的板卡位置检测方法,其特征在于,所述计算所述轮廓的边缘点坐标的步骤包括:获取所述轮廓的外接矩阵的最值点坐标,将所述最值点坐标作为所述轮廓的边缘点坐标。
- 根据权利要求1所述的板卡位置检测方法,其特征在于,所述获取所述最大值对应的连通区域的中心点坐标的步骤包括:获取所述最大值对应的连通区域的轮廓;计算所述轮廓的边角坐标和所述轮廓的长度与宽度;根据所述边角坐标和所述长度与宽度,获取所述中心点坐标。
- 根据权利要求1所述的板卡位置检测方法,其特征在于,所述计算所述形态学图像中的各连通区域的面积,并获取所述面积中的最大值的步骤包括:分别获取所述形态学图像中各连通区域的轮廓;对所述轮廓进行面积计算,得到所述轮廓的面积。
- 根据权利要求1至5任意一项所述的板卡位置检测方法,其特征在于,所述数学形态学运算的运算内核为矩形核。
- 一种板卡位置检测装置,其特征在于,包括:图像获取模块,用于拍摄待检测位置的图像,获取所述图像的灰度图;图像处理模块,用于对所述灰度图进行边缘检测处理,得到所述灰度图的边缘图;并获取所述边缘图的包含若干连通区域的形态学图像;以及计算所述形态学图像中的各连通区域的面积;判断模块,用于获取所述面积中的最大值,并判断所述最大值是否大于或等于预设面积阈值;位置获取模块,用于在所述判断模块的判断结果为是时,获取所述最大值对应的连通区域的中心点坐标,并将所述中心点坐标作为所述板卡的中心点坐标。
- 根据权利要求7所述的板卡位置检测装置,其特征在于,所述位置获取模块包括:第一轮廓获取模块,用于获取所述最大值对应的连通区域的轮廓;第一坐标获取模块,用于计算所述轮廓的边缘点坐标;并根据所述边缘点坐标,获取所述中心点坐标。
- 根据权利要求7所述的板卡位置检测装置,其特征在于,所述位置获取模块包括:第一轮廓获取模块,用于获取所述最大值对应的连通区域的轮廓;第二坐标获取模块,用于计算所述轮廓的边角坐标和所述轮廓的长度与宽度;并根据所述边角坐标和所述长度与宽度,获取所述中心点坐标。
- 根据权利要求7至9任意一项所述的板卡位置检测装置,其特征在于,所述图像处理模块包括:第二轮廓获取模块,用于获取所述形态学图像中各连通区域的轮廓;面积获取模块,用于对所述轮廓进行面积计算,得到所述轮廓的面积。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510697415.3 | 2015-10-21 | ||
CN201510697415.3A CN105352437B (zh) | 2015-10-21 | 2015-10-21 | 板卡位置检测方法及装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017067342A1 true WO2017067342A1 (zh) | 2017-04-27 |
Family
ID=55328448
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2016/098222 WO2017067342A1 (zh) | 2015-10-21 | 2016-09-06 | 板卡位置检测方法及装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN105352437B (zh) |
WO (1) | WO2017067342A1 (zh) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114047123A (zh) * | 2021-10-09 | 2022-02-15 | 江苏芯亿达电子科技有限公司 | 一种用于集成板生产缺陷检测方法及检测系统 |
CN114494060A (zh) * | 2022-01-20 | 2022-05-13 | 齐鲁工业大学 | 一种基于Opencv的菲涅尔透镜中心定位方法 |
CN115086508A (zh) * | 2021-03-11 | 2022-09-20 | 盛视科技股份有限公司 | 集装箱检测方法及装置、安防车载系统及安防监控系统 |
CN116416181A (zh) * | 2021-12-29 | 2023-07-11 | 广州镭晨智能装备科技有限公司 | 一种插针缺陷的检测方法、检测设备和存储介质 |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105352437B (zh) * | 2015-10-21 | 2019-05-07 | 广州视源电子科技股份有限公司 | 板卡位置检测方法及装置 |
CN105975910A (zh) * | 2016-04-27 | 2016-09-28 | 梧州市自动化技术研究开发院 | 一种用于对运动物体进行视频识别的技术处理方法及其系统 |
CN106228546B (zh) * | 2016-07-13 | 2019-01-29 | 广州视源电子科技股份有限公司 | 一种板卡的检测方法及装置 |
CN106556352A (zh) * | 2016-11-30 | 2017-04-05 | 英业达科技有限公司 | 一种自动检测装置、系统及方法 |
CN109543665B (zh) * | 2017-09-22 | 2020-10-16 | 凌云光技术集团有限责任公司 | 图像定位方法及装置 |
CN108918093B (zh) * | 2018-05-23 | 2020-08-04 | 精锐视觉智能科技(深圳)有限公司 | 一种滤光片镜面缺陷检测方法、装置及终端设备 |
CN109190616B (zh) * | 2018-08-03 | 2021-10-15 | 东北大学 | 一种基于特征识别的热轧钢板在线视觉跟踪方法 |
CN111102917A (zh) * | 2018-10-26 | 2020-05-05 | 得力富企业股份有限公司 | 电路板的形心侦测设备 |
CN113034531B (zh) * | 2021-04-02 | 2024-09-17 | 广州绿怡信息科技有限公司 | 设备摆放检测方法及装置 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003086919A (ja) * | 2001-09-07 | 2003-03-20 | Sun Tec Kk | パターン検査装置 |
CN102721364A (zh) * | 2011-03-30 | 2012-10-10 | 比亚迪股份有限公司 | 一种工件的定位方法及其装置 |
CN103235939A (zh) * | 2013-05-08 | 2013-08-07 | 哈尔滨工业大学 | 基于机器视觉的基准点定位方法 |
CN104297254A (zh) * | 2014-10-08 | 2015-01-21 | 华南理工大学 | 一种基于混合法的印刷电路板缺陷检测方法及系统 |
CN104484871A (zh) * | 2014-11-27 | 2015-04-01 | 小米科技有限责任公司 | 边缘提取方法和装置 |
CN105352437A (zh) * | 2015-10-21 | 2016-02-24 | 广州视源电子科技股份有限公司 | 板卡位置检测方法及装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005164579A (ja) * | 2004-10-29 | 2005-06-23 | Alps Electric Co Ltd | 境界線の位置の検出方法およびこの検出方法を用いた磁気ヘッドの位置決め方法および位置決め装置 |
CN102163284B (zh) * | 2011-04-11 | 2013-02-27 | 西安电子科技大学 | 面向中文环境的复杂场景文本定位方法 |
CN104408722A (zh) * | 2014-11-26 | 2015-03-11 | 南京工业大学 | 一种用于pcb视觉定位的检测方法 |
CN104794421B (zh) * | 2015-04-29 | 2017-09-26 | 华中科技大学 | 一种qr码定位及识别方法 |
-
2015
- 2015-10-21 CN CN201510697415.3A patent/CN105352437B/zh active Active
-
2016
- 2016-09-06 WO PCT/CN2016/098222 patent/WO2017067342A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003086919A (ja) * | 2001-09-07 | 2003-03-20 | Sun Tec Kk | パターン検査装置 |
CN102721364A (zh) * | 2011-03-30 | 2012-10-10 | 比亚迪股份有限公司 | 一种工件的定位方法及其装置 |
CN103235939A (zh) * | 2013-05-08 | 2013-08-07 | 哈尔滨工业大学 | 基于机器视觉的基准点定位方法 |
CN104297254A (zh) * | 2014-10-08 | 2015-01-21 | 华南理工大学 | 一种基于混合法的印刷电路板缺陷检测方法及系统 |
CN104484871A (zh) * | 2014-11-27 | 2015-04-01 | 小米科技有限责任公司 | 边缘提取方法和装置 |
CN105352437A (zh) * | 2015-10-21 | 2016-02-24 | 广州视源电子科技股份有限公司 | 板卡位置检测方法及装置 |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115086508A (zh) * | 2021-03-11 | 2022-09-20 | 盛视科技股份有限公司 | 集装箱检测方法及装置、安防车载系统及安防监控系统 |
CN114047123A (zh) * | 2021-10-09 | 2022-02-15 | 江苏芯亿达电子科技有限公司 | 一种用于集成板生产缺陷检测方法及检测系统 |
CN114047123B (zh) * | 2021-10-09 | 2024-04-09 | 无锡芯羿达微电子有限公司 | 一种用于集成板生产缺陷检测方法及检测系统 |
CN116416181A (zh) * | 2021-12-29 | 2023-07-11 | 广州镭晨智能装备科技有限公司 | 一种插针缺陷的检测方法、检测设备和存储介质 |
CN116416181B (zh) * | 2021-12-29 | 2024-05-31 | 广州镭晨智能装备科技有限公司 | 一种插针缺陷的检测方法、检测设备和存储介质 |
CN114494060A (zh) * | 2022-01-20 | 2022-05-13 | 齐鲁工业大学 | 一种基于Opencv的菲涅尔透镜中心定位方法 |
CN114494060B (zh) * | 2022-01-20 | 2024-02-23 | 齐鲁工业大学 | 一种基于Opencv的菲涅尔透镜中心定位方法 |
Also Published As
Publication number | Publication date |
---|---|
CN105352437A (zh) | 2016-02-24 |
CN105352437B (zh) | 2019-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017067342A1 (zh) | 板卡位置检测方法及装置 | |
CN108460757B (zh) | 一种手机TFT-LCD屏Mura缺陷在线自动检测方法 | |
US7936926B2 (en) | Apparatus, method, and program for face feature point detection | |
WO2020253827A1 (zh) | 评估图像采集精度的方法及装置、电子设备和存储介质 | |
Ardhy et al. | Development of SBC based machine-vision system for PCB board assembly automatic optical inspection | |
JP2007285754A (ja) | 欠陥検出方法および欠陥検出装置 | |
JP2007172397A (ja) | エッジ勾配検出方法、シミ欠陥検出方法、エッジ勾配検出装置、シミ欠陥検出装置 | |
CN107610085A (zh) | 一种基于计算机视觉的焊点缺陷检测系统 | |
CN108918093B (zh) | 一种滤光片镜面缺陷检测方法、装置及终端设备 | |
KR101677070B1 (ko) | 형태학적 영상 처리와 레이블링을 이용한 얼룩 결함 자동 검출 시스템 및 방법 | |
JP2005345290A (ja) | 筋状欠陥検出方法及び装置 | |
JP2006258713A (ja) | シミ欠陥検出方法及び装置 | |
JP2005165387A (ja) | 画面のスジ欠陥検出方法及び装置並びに表示装置 | |
JP2019200775A (ja) | 表面欠陥検査装置及び表面欠陥検査方法 | |
CN111816579B (zh) | 用于检测被助焊剂残留物污染的芯片的方法和装置 | |
JP2003167529A (ja) | 画面欠陥検出方法及び装置並びに画面欠陥検出のためのプログラム | |
JP2008014842A (ja) | シミ欠陥検出方法及び装置 | |
JP2008171142A (ja) | シミ欠陥検出方法及び装置 | |
JP2007285868A (ja) | 輝度勾配検出方法、欠陥検出方法、輝度勾配検出装置および欠陥検出装置 | |
JP2006145228A (ja) | ムラ欠陥検出方法及び装置 | |
CN111815705A (zh) | 激光跟踪仪滤光保护镜片污染识别方法、装置及电子设备 | |
KR102203441B1 (ko) | 전자부품 분류장치 및 그 방법 | |
TWI510776B (zh) | 玻璃氣泡瑕疵檢測處理方法 | |
KR20120036145A (ko) | 레이저 비전 시스템을 이용한 용접 조인트의 특징점 검출 방법 | |
JP2006133055A (ja) | ムラ欠陥検出方法及び装置、空間フィルタ、ムラ欠陥検査システム並びにムラ欠陥検出方法のプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16856775 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 16856775 Country of ref document: EP Kind code of ref document: A1 |