WO2017080295A1 - 元件的定位方法和系统 - Google Patents

元件的定位方法和系统 Download PDF

Info

Publication number
WO2017080295A1
WO2017080295A1 PCT/CN2016/098226 CN2016098226W WO2017080295A1 WO 2017080295 A1 WO2017080295 A1 WO 2017080295A1 CN 2016098226 W CN2016098226 W CN 2016098226W WO 2017080295 A1 WO2017080295 A1 WO 2017080295A1
Authority
WO
WIPO (PCT)
Prior art keywords
tested
picture
unit
image
sub
Prior art date
Application number
PCT/CN2016/098226
Other languages
English (en)
French (fr)
Inventor
罗汉杰
Original Assignee
广州视源电子科技股份有限公司
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 广州视源电子科技股份有限公司 filed Critical 广州视源电子科技股份有限公司
Publication of WO2017080295A1 publication Critical patent/WO2017080295A1/zh

Links

Images

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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • 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 present invention relates to the field of automated optical inspection, and more particularly to methods and systems for positioning components.
  • PCB printed circuit board
  • AOI Automatic Optical Inspection
  • the AOI system optically obtains the image information of the object to be tested, and then uses the image processing algorithm to perform the object to be measured. Detection, check for foreign objects, defects, etc.
  • the AOI system used in PCB board detection it is necessary to perform error, leak, and reverse detection on the components above the board.
  • the components may be displaced, that is, the position of the components on the board is not fixed, so before the components are tested for errors, leaks, and reverses, Position the component first.
  • the component In the current AOI system, the component is generally only detected at a fixed position, and the component is not positioned. Therefore, when detecting a component with a high offset, such as a diode or a capacitor, it may be tested. There are no components in the area that cause detection errors.
  • a method for locating components includes the following steps:
  • the matching degree is less than the first threshold, it is determined that the position of the subgraph is the position of the component to be tested in the map to be tested.
  • a component positioning system comprising the following units:
  • a first acquiring unit configured to acquire a template image of the component to be tested and a to-be-tested image actually taken by the component to be tested;
  • a second acquiring unit configured to acquire a color distribution descriptor of the template image
  • a third acquiring unit configured to obtain, in the to-be-tested image, a picture of the same size as the template image as a sub-picture of the to-be-tested picture;
  • a fourth obtaining unit configured to acquire a color distribution descriptor of the subgraph
  • a calculating unit configured to calculate a matching degree between the color distribution descriptor of the template image and the color distribution descriptor of the sub-picture
  • the determining unit is configured to determine, when the matching degree is less than the first threshold, a position where the subgraph is located as a position of the device to be tested in the to-be-tested image.
  • the template of the device to be tested and the image to be tested of the device to be tested are used to obtain a color distribution descriptor of the template image, and the grayscale image of the template image and the grayscale image of the image to be tested are obtained.
  • Perform matching to obtain a matching matrix, and take a pixel point in the to-be-measured image corresponding to the element with the smallest element value as a vertex of the sub-graph, obtain the sub-picture and its color distribution descriptor in the to-be-tested image, and then according to the template image
  • the degree of matching between the color distribution descriptor and the color distribution descriptor of the subgraph determines the position of the component to be tested in the map to be tested. According to the image of the known component, the position of the component can be positioned accurately in the image to be tested, and the positioning is accurate, which provides an important basis for the component to perform error, leak, and reverse detection.
  • Figure 1 is a flow chart of a method of positioning components in one of the embodiments
  • Figure 2 is a flow chart of a method of positioning components in one of the embodiments
  • Figure 3 is a flow chart of a method of positioning components in one of the embodiments
  • Figure 4 is a schematic view of the positioning system of the component corresponding to Figure 1;
  • Figure 5 is a partial schematic view of a positioning system of components in one of the embodiments.
  • Figure 6 is a partial schematic view of a positioning system of components in one of the embodiments.
  • Figure 7 is a partial schematic view of a positioning system of components in one of the embodiments.
  • Figure 8 is a partial schematic view of a positioning system of components in one of the embodiments.
  • Figure 9 is a schematic illustration of a positioning system for components in one of the embodiments.
  • Figure 10 is a schematic illustration of a positioning system for components in one of the embodiments.
  • FIG. 1 an embodiment of a method of locating an element of the present invention is shown.
  • the positioning method of the components in this embodiment includes the following steps:
  • Step S110 acquiring a template image of the component to be tested and a to-be-tested image actually taken by the component to be tested;
  • the component to be tested can be an electronic component on the PCB, such as a resistor, an inductor, a capacitor, etc.; the template image only includes image information of the component to be tested; the to-be-tested image is a PCB board image including the component to be tested, and is included The PCB of the measuring component is captured;
  • Step S120 Acquire a color distribution descriptor of the template image
  • the color distribution descriptor of the template image can be used to represent the color information of the template image
  • Step S130 Obtain a picture of the same size as the template image in the to-be-tested picture as a sub-picture of the to-be-tested picture;
  • Step S140 Acquire a color distribution descriptor of the subgraph
  • the color distribution descriptor of the subgraph can be used to represent the color information of the subgraph
  • Step S150 calculating a matching degree between the color distribution descriptor of the template image and the color distribution descriptor of the sub-picture;
  • the matching degree is calculated according to the color distribution descriptor of the template image and the color distribution descriptor of the sub-picture, and the higher the value, the more similar the color information of the template picture and the color information of the sub-picture are;
  • Step S160 If the matching degree is less than the first threshold, determine that the position where the subgraph is located is the position of the component to be tested in the to-be-tested image;
  • the first threshold is used to define the degree of matching, thereby determining whether the position of the subgraph is the position of the component to be tested in the map to be tested.
  • the positioning method of the component described in this embodiment is to use a template diagram of the component to be tested and a component to be tested
  • the image to be tested obtains the color distribution descriptor of the template image, and matches the grayscale image of the template image with the grayscale image of the image to be tested to obtain a matching matrix, and takes the point with the smallest element value as the minimum coordinate point.
  • the subgraph and its color distribution descriptor are obtained in the map, and the position of the device to be tested in the map to be tested is determined according to the matching degree between the color distribution descriptor of the template graph and the color distribution descriptor of the subgraph.
  • the position of the component can be positioned accurately in the image to be tested, and the positioning is accurate, which provides an important basis for the component to perform error, leak, and reverse detection.
  • the step of obtaining a color distribution descriptor of the template map comprises the steps of:
  • n is a positive integer, and average the R, G, and B channels of the RGB color space of each unit graph to obtain an average of each unit graph.
  • the step of obtaining the color distribution descriptor of the subgraph includes the following steps:
  • the color description distributor obtained in this manner can be compared. Goodly cover the color information of the template or subgraph.
  • a template map of the component to be tested is obtained, and the template map is defined as
  • h model is the height of the template
  • w model is the width of the template
  • the map to be tested is defined as
  • h object is the height of the graph to be tested
  • w object is the width of the graph to be tested.
  • the template map I model (x, y) is equally divided into n ⁇ n unit maps S(x', y') of the same size, where x' ⁇ [0,n), y' ⁇ [0,n), x', y', n are positive integers, for each unit map S(x', y'), respectively, for its RGB three channels Take the average, so each unit map S(x', y') gets an average color:
  • RGB(x', y') ⁇ AVG(R(x', y')), AVG(R(x', y')), AVG(R(x', y')) ⁇
  • the color space is transformed for each unit map S(x, y), and RGB(x', y') represented in the RGB color space is converted into a form represented in the YCbCr color space:
  • YCbCr(x', y') ⁇ Y(x', y'), Cb(x', y'), Cr(x', y') ⁇
  • CLD model ⁇ (x',y'),(x',y'),(x',y') ⁇
  • the CLD model here is the color distribution descriptor of the template image.
  • the value of n is generally 8 or other positive integers.
  • This calculation method is also used when obtaining the color distribution descriptor of the sub-image I object_sub , and the value of m may be the same as or different from the color distribution descriptor of the template image, and the value of m is also 8
  • Sub-image I object_sub is divided into m ⁇ m unit maps of the same size, and the R, G, and B channels of the RGB color space of each unit map are respectively averaged to obtain an average color of each unit map;
  • the RGB color space of each unit map is converted into a YCbCr color space, and a conversion map having three channels of Y, Cb, and Cr is obtained, and three channels of the conversion graph are discrete cosine transformed to obtain the sub-image I object_sub
  • the color distribution descriptor CLD object_sub the expression is:
  • CLD object_sub ⁇ (x',y'),(x',y'),(x',y') ⁇
  • the step of acquiring a picture of the same size as the template image as a sub-picture of the to-be-tested picture in the to-be-tested view includes the following steps:
  • Step S131 acquiring a grayscale image of the template image and a grayscale image of the to-be-measured image, and obtaining a matching matrix of the grayscale image of the template image in the grayscale image of the to-be-tested image, wherein each element in the matching matrix is Corresponding to the pixel points in the graph to be tested;
  • Step S132 Perform screening on the elements in the matching matrix to obtain a set of elements in the matching matrix that are smaller than the second threshold.
  • the second threshold can be freely set according to the actual condition of the component
  • Step S133 In the set of elements smaller than the second threshold, select a point in the to-be-measured map corresponding to the element with the smallest element value, and obtain a picture of the same size as the template picture in the to-be-tested picture as the sub-picture of the to-be-tested picture.
  • the lateral edge of the subgraph is parallel to the lateral edge of the graph to be tested
  • the longitudinal edge of the subgraph is parallel to the longitudinal edge of the graph to be tested
  • one vertex of the subgraph is the pixel in the graph to be tested corresponding to the element with the smallest element value ;
  • the step of acquiring the grayscale image of the template image and the grayscale image of the image to be tested, and obtaining the matching matrix of the grayscale image of the template image in the grayscale image of the image to be tested includes the following steps:
  • the matching matrix obtained by the above method can well reflect the similarity degree of each pixel in the graph to be compared with the template graph.
  • the I' model and the I' object of the I model and the I object are respectively obtained, and the normalized squared difference matching method is used, and the I' model is used as a convolution kernel, and the I' object is convoluted and obtained.
  • the matching matrix R of the I' model in the I' object the specific formula is as follows:
  • Each element R(x, y) in the matching matrix R corresponds to the gray value of a certain pixel in the I' object , and also corresponds to the corresponding pixel in the I object .
  • the elements in the matching matrix are filtered to obtain a small matching matrix.
  • the step of collecting the elements of the first threshold includes the following steps:
  • the matching matrix R is filtered to obtain a set of elements in which the element value is smaller than the second threshold r threshold . If the element set is an empty set, it is determined that the element to be tested does not appear in the map to be tested.
  • Matched filter matrix R for the elements, less than the second threshold r threshold selected set of elements the matrix can be avoided matching elements other than or equal to a second threshold value for subsequent operations, even if the subsequent operations of these elements can not be Obtaining the appropriate results, thereby reducing the workload; and in the case of an empty set of elements, it can be directly determined that the device under test does not appear in the map to be tested, and no further operations are required.
  • the way to obtain the subgraph is:
  • the pixel in the to-be-measured image corresponding to the element with the smallest element value is selected, and the image of the same size as the template image is obtained as the sub-graph of the to-be-tested image in the image to be tested, where
  • the lateral edge of the subgraph is parallel to the lateral edge of the graph to be tested, and the longitudinal edge of the subgraph is parallel to the longitudinal edge of the graph to be tested.
  • One vertex of the subgraph is the pixel point in the graph to be tested corresponding to the element with the smallest element value.
  • a sub-graph having the same size as the template image and parallel to the edge position of the template image can be obtained in the image to be tested, and as long as the same component as the template image exists in the to-be-tested image, the sub-graph can be used. Position to determine the position of the component under test in the map to be tested.
  • the range of x best and y best is [0, w object –w model ], [0, h object –h model ], and the maximum value of the subgraph obtained by taking the best matching point as the vertex is w
  • the maximum value of object and y is h object , so the obtained subgraph is always within the bounds of the graph to be tested.
  • calculating a matching degree between the color distribution descriptor of the template image and the color distribution descriptor of the sub-picture; if the matching degree is less than the first threshold, determining that the position of the sub-picture is the component to be tested is to be tested The steps in the figure are as follows:
  • the matching degree d between the two is calculated, and the formula is as follows:
  • the position of the subgraph I object_sub is the position of the device to be tested in the map to be tested.
  • the positions of the components are not necessarily at the predetermined correct position, the sub-picture is determined by the best matching point, and the image position information of the component is utilized; the color description distributor and the sub-picture of the comparison template image are compared.
  • the color description distributor uses the image color information of the component, and the solution of the present invention combines the shape information and the color information of the image, which is more accurate than the ordinary positioning algorithm.
  • the step of calculating the degree of matching between the color distribution descriptor of the template map and the color distribution descriptor of the sub-graph includes the following steps:
  • the matching degree is greater than or equal to the first threshold, the elements in the element set smaller than the second threshold corresponding to the vertex are cleared, and returned to the element set smaller than the second threshold, and the element with the smallest element value is selected to be in the to-be-tested picture.
  • the corresponding element whose matching degree is greater than or equal to the first threshold is cleared, and is re-selected from the set of elements; in the actual positioning process, the element value of an element may be the smallest, and the matching degree is not This case excludes this situation where appropriate.
  • the matching degree d is greater than or equal to the first threshold d threshold , it indicates that the subgraph corresponding to the best matching point (x best , y best ) has insufficient matching degree in the color space, and the best matching is in the element set.
  • the element in the set of elements corresponding to the point corresponding to the first threshold is cleared, and the set of elements after the corresponding element is cleared as a new set of elements.
  • the new element set select the point in the map to be tested corresponding to the element with the smallest element value as the best matching point, and select a new sub-picture according to the above method to obtain the color description distributor of the new sub-picture, and then calculate The new matching degree between the color distribution descriptor of the template image and the color distribution descriptor of the new subgraph is finally compared with the first threshold d threshold . If the new matching degree is smaller than the first threshold d threshold , the subgraph is determined.
  • the position of the component to be tested is in the map to be tested; if greater than or equal to the first threshold d threshold , the elements in the set of elements corresponding to the second threshold are correspondingly cleared again; repeat the above steps until Determining where the component to be tested is located in the map to be tested, or the set of elements becomes an empty set;
  • the judgment processing can also be performed in the case where all the elements in the set of elements are cleared.
  • the invention provides a method for positioning components. According to the image of the known components, the position of the components in the map to be tested is accurately positioned, which provides an important basis for detecting faults, leaks and reverses of the components. The problem that the component is not in the detection area due to the displacement is avoided, and the accuracy of the component detection result is enhanced, and the invention combines the shape information and the color information of the image, which is more accurate than the ordinary positioning algorithm.
  • the present invention also provides a positioning system for components, and an embodiment of the positioning system of the components of the present invention will be described in detail below.
  • the positioning system of the component in this embodiment includes a first obtaining unit 210, a second obtaining unit 220, a third obtaining unit 230, a fourth obtaining unit 240, a calculating unit 250, and a determining unit 260, wherein:
  • a first acquiring unit 210 configured to acquire a template image of the component to be tested and a to-be-tested image actually taken by the component to be tested;
  • a second obtaining unit 220 configured to acquire a color distribution descriptor of the template image
  • the third obtaining unit 230 is configured to obtain, in the to-be-tested image, a picture of the same size as the template image as a sub-picture of the to-be-tested picture;
  • a fourth obtaining unit 240 configured to acquire a color distribution descriptor of the subgraph
  • the calculating unit 250 is configured to calculate a matching degree between the color distribution descriptor of the template image and the color distribution descriptor of the sub-picture;
  • the determining unit 260 is configured to determine, when the matching degree is less than the first threshold, that the position of the subgraph is to be Measure the position of the component in the map to be tested.
  • the second obtaining unit 220 includes a first sub-picture unit 221, a first value unit 222, and a first conversion unit 223;
  • the first sub-picture unit 221 is configured to divide the template picture into n ⁇ n unit maps of the same size, where n is a positive integer;
  • the first value unit 222 is configured to average the R, G, and B channels of the RGB color space of each unit map to obtain an average color of each unit map;
  • the first converting unit 223 is configured to convert the RGB color space of each unit map into a YCbCr color space, obtain a conversion map having three channels of Y, Cb, and Cr, and perform discrete cosine transform on the three channels of the conversion graph, Obtaining a color distribution descriptor of the template map;
  • the fourth obtaining unit 240 includes a second sub-picture unit 241, a second value unit 242, and a second conversion unit 243;
  • a second sub-picture unit 241 configured to divide the sub-picture into m ⁇ m unit maps of the same size, where m is a positive integer;
  • the second value unit 242 is configured to average the R, G, and B channels of the RGB color space of each unit map to obtain an average color of each unit map;
  • the second converting unit 243 is configured to convert the RGB color space of each unit map into a YCbCr color space, obtain a conversion map having three channels of Y, Cb, and Cr, and perform discrete cosine transform on the three channels of the conversion graph, Obtain the color distribution descriptor of the subgraph.
  • the third obtaining unit 230 includes the following units:
  • the matrix obtaining unit 231 is configured to obtain a grayscale image of the template image and a grayscale image of the to-be-measured image, and obtain a matching matrix of the grayscale image of the template image in the grayscale image of the to-be-tested image, where the matching matrix Each element corresponds to a pixel in the graph to be tested;
  • the filtering unit 232 is configured to filter elements in the matching matrix to obtain a set of elements in the matching matrix that are smaller than the second threshold;
  • a sub-graph obtaining unit 233 configured to: after the filtering unit obtains the set of elements that are smaller than the second threshold in the matching matrix, select, in the set of elements that is smaller than the second threshold, the pixel points in the to-be-measured image corresponding to the element with the smallest element value, And obtaining a picture of the same size as the template picture in the to-be-tested picture as a sub-picture of the to-be-tested picture, wherein, the lateral edge of the subgraph is parallel to the lateral edge of the graph to be tested, the longitudinal edge of the subgraph is parallel to the longitudinal edge of the graph to be tested, and one vertex of the subgraph is the pixel in the graph to be tested corresponding to the element with the smallest element value .
  • the positioning system of the component further includes a second determining unit 270, configured to determine, when the set of elements smaller than the second threshold in the obtained matching matrix is an empty set, determine that the device to be tested is not Appears in the graph to be tested.
  • the positioning system of the component further includes a clearing unit 280.
  • the clearing unit 280 is configured to: when the matching degree is greater than or equal to the first threshold, clear an element in the set of elements that is smaller than the second threshold corresponding to the vertex;
  • the sub-graph obtaining unit 233 is further configured to: after the element is cleared by the clearing unit, select a pixel point in the to-be-measured image corresponding to the element with the smallest element value in the set of elements that is smaller than the second threshold, and in the to-be-tested image Obtaining a picture of the same size as the template image as a sub-picture of the image to be tested, wherein the lateral edge of the sub-picture is parallel to the lateral edge of the image to be tested, and the longitudinal edge of the sub-picture is parallel to the longitudinal edge of the image to be tested, and one of the sub-pictures The vertex is the pixel point in the graph to be tested corresponding to the element with the smallest element value.
  • some units are cycled until it is determined that the device under test is in the position in the map to be tested, or the set of elements becomes an empty set.
  • the second determining unit 270 is further configured to determine that the device under test does not appear in the map to be tested when all the elements in the set of elements smaller than the second threshold are cleared.
  • the positioning system of the component of the present invention corresponds one-to-one with the positioning method of the component of the present invention, and the technical features and advantageous effects of the embodiment of the positioning method of the above-described component are applicable to the embodiment of the positioning system of the component.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Quality & Reliability (AREA)

Abstract

一种元件的定位方法和系统，利用待测元件的模板图和待测元件所在的待测图，获得模板图的颜色分布描述符（S120），将模板图的灰度图和待测图的灰度图进行匹配得到匹配矩阵（S131），并以其中元素值最小的元素对应的待测图中的像素点作为子图的一个顶点，在待测图中获取子图及其颜色分布描述符（S140），再根据模板图的颜色分布描述符和子图的颜色分布描述符的匹配度来确定待测元件在待测图中的位置（S150）。以此方案可根据已知元件的图像，在待测图中定位元件的位置，定位准确，为对元件进行错，漏，反等检测提供了重要依据。

Description

元件的定位方法和系统 技术领域
本发明涉及自动光学检测领域，特别是涉及元件的定位方法和系统。
背景技术
当前，对PCB(印制电路板)板卡进行检测，使用较多的是AOI(自动光学检测)系统，其是利用光学方式取得待测物体的图像信息，然后使用图像处理算法对待测物体进行检测，检查出异物，瑕疵等错误。对于使用在PCB板卡检测的AOI系统中，需要对板卡上面的元件进行错，漏，反等检测。然而在插入的过程，或者上焊锡的过程中，元件可能会发生位移，即元件在板卡上面的位置并非都是固定不变的，所以在对元件进行错，漏，反等检测前，需要先对元件进行定位。
现在的AOI系统中，一般只会在固定的位置检测元件，而不会对元件进行定位，如此，对于偏移量较高的元件，如二极管，电容等，进行检测时，可能会因为待测区域中没有元件而导致检测错误。
发明内容
基于此，有必要针对元件发生位移导致检测错误的问题，提供一种元件的定位方法和系统。
一种元件的定位方法，包括以下步骤：
获取待测元件的模板图和对待测元件实际拍摄的待测图；
获取模板图的颜色分布描述符；
在待测图中获取与模板图相同大小的图片作为待测图的子图；
获取子图的颜色分布描述符；
计算模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度；
若匹配度小于第一阈值，则判定子图所在的位置为待测元件在待测图中所在的位置。
一种元件的定位系统，包括以下单元：
第一获取单元，用于获取待测元件的模板图和对待测元件实际拍摄的待测图；
第二获取单元，用于获取模板图的颜色分布描述符；
第三获取单元，用于在待测图中获取与模板图相同大小的图片作为待测图的子图；
第四获取单元，用于获取子图的颜色分布描述符；
计算单元，用于计算模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度；
判断单元，用于在匹配度小于第一阈值时，判定子图所在的位置为待测元件在待测图中所在的位置。
根据上述本发明的方案，其是利用待测元件的模板图和待测元件所在的待测图，获得模板图的颜色分布描述符，将模板图的灰度图和待测图的灰度图进行匹配得到匹配矩阵，并以其中元素值最小的元素对应的待测图中的像素点作为子图的一个顶点，在待测图中获取子图及其颜色分布描述符，再根据模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度来确定待测元件在待测图中的位置。以此方案可根据已知元件的图像，在待测图中定位元件的位置，定位准确，为对元件进行错，漏，反等检测提供了重要依据。
附图说明
图1是其中一个实施例中元件的定位方法的流程图；
图2是其中一个实施例中元件的定位方法的流程图；
图3是其中一个实施例中元件的定位方法的流程图；
图4是图1对应的元件的定位系统的示意图；
图5是其中一个实施例中元件的定位系统的部分示意图；
图6是其中一个实施例中元件的定位系统的部分示意图；
图7是其中一个实施例中元件的定位系统的部分示意图；
图8是其中一个实施例中元件的定位系统的部分示意图；
图9是其中一个实施例中元件的定位系统的示意图；
图10是其中一个实施例中元件的定位系统的示意图。
具体实施方式
为使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步的详细说明。应当理解，此处所描述的具体实施方式仅仅用以解释本发明，并不限定本发明的保护范围。
参见图1所示，为本发明的元件的定位方法的实施例。该实施例中的元件的定位方法包括如下步骤：
步骤S110：获取待测元件的模板图和对待测元件实际拍摄的待测图；
待测元件可以为PCB板上的电子元器件，如电阻、电感、电容等；模板图中只包括待测元件的图像信息；待测图是包括待测元件的PCB板图像，是对包括待测元件的PCB板拍摄得到的；
步骤S120：获取模板图的颜色分布描述符；
模板图的颜色分布描述符可以用来表示模板图的颜色信息；
步骤S130：在待测图中获取与模板图相同大小的图片作为待测图的子图；
步骤S140：获取子图的颜色分布描述符；
子图的颜色分布描述符可以用来表示子图的颜色信息；
步骤S150：计算模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度；
匹配度是根据模板图的颜色分布描述符和子图的颜色分布描述符计算的，其值越高，表示模板图的颜色信息与子图的颜色信息越相似；
步骤S160：若匹配度小于第一阈值，则判定子图所在的位置为待测元件在待测图中所在的位置；
第一阈值用来界定匹配度，从而确定子图的位置是否为待测元件在待测图中所在的位置。
本实施方式所述的元件的定位方法，是利用待测元件的模板图和待测元件 所在的待测图，获得模板图的颜色分布描述符，将模板图的灰度图和待测图的灰度图进行匹配得到匹配矩阵，并以其中元素值最小的点作为最小坐标点在待测图中获取子图及其颜色分布描述符，再根据模板图的颜色分布描述符和所述子图的颜色分布描述符之间的匹配度来确定待测元件在待测图中的位置。以此方案可根据已知元件的图像，在待测图中定位元件的位置，定位准确，为对元件进行错，漏，反等检测提供了重要依据。
在其中一个实施例中，获取模板图的颜色分布描述符的步骤包括以下步骤：
将模板图均分成n×n个相同大小的单元图，n为正整数，对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的Y、Cb、Cr三个通道进行离散余弦变换，获得模板图的颜色分布描述符；
获取子图的颜色分布描述符的步骤包括以下步骤：
将子图均分成m×m相同大小的单元图，m为正整数，对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的Y、Cb、Cr三个通道进行离散余弦变换，获得子图的颜色分布描述符。
通过将模板图或子图分割成单元图，获取每个单元图的平均颜色并进行颜色空间转换和离散余弦变换，获得模板图的颜色描述分布符，此种方式获得的颜色描述分布符可以较好地全面覆盖模板图或子图的颜色信息。
优选的，获取待测元件的模板图，模板图定义为
Imodel(x,y)，x∈[0,wmodel)，y∈[0,hmodel)，
hmodel为模版图的高，wmodel为模版图的宽；
获取要进行的定位对待测元件实际拍摄的待测图，待测图定义为
Iobject(x,y),x∈[0,wobject)，y∈[0,hobject)，
hobject为待测图的高，wobject为待测图的宽。
以上述获取的模板图Imodel(x,y)为例，将模板图Imodel(x,y)等分成n×n个相同大小的单元图S(x’,y’)，其中，x’∈[0,n)，y’∈[0,n)，x’、y’、n均为正整数，对每一个单元图S(x’,y’)，分别对它的RGB三个通道取平均值，所以每一个单元图S(x’,y’)都获得一个平均颜色：
RGB(x’,y’)＝{AVG(R(x’,y’)),AVG(R(x’,y’)),AVG(R(x’,y’))}
对每一个单元图S(x,y)转变颜色空间，将在RGB颜色空间中表示的RGB(x’,y’)转变为在YCbCr颜色空间中表示的形式：
YCbCr(x’,y’)＝{Y(x’,y’),Cb(x’,y’),Cr(x’,y’)}
获得一个n×n大小的，具有YCbCr三通道的图IYCbCr(x’,y’)；
对图IYCbCr(x’,y’)每一个通道进行离散余弦变换，获得3个n×n大小的矩阵
CLDmodel＝{(x’,y’),(x’,y’),(x’,y’)}
这里的CLDmodel即为模板图的颜色分布描述符。
根据计算经验，一般n取值为8，也可为其他正整数。在获取子图Iobject_sub的颜色分布描述符时也是采用这种计算方法，而且m的取值与计算模板图的颜色分布描述符时的n可以相同，也可以不同，一般m取值也为8，将子图Iobject_sub分成m×m个相同大小的单元图，对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对所述转换图的三个通道进行离散余弦变换，获得所述子图Iobject_sub的颜色分布描述符CLDobject_sub,表达式为：
CLDobject_sub＝{(x’,y’),(x’,y’),(x’,y’)}
在其中一个实施例中，如图2所示，在待测图中获取与模板图相同大小的图片作为待测图的子图的步骤包括以下步骤：
步骤S131：获取模板图的灰度图和待测图的灰度图，并获得模板图的灰度图在待测图的灰度图中的匹配矩阵，其中，匹配矩阵中的每个元素均对应于待测图中的像素点；
步骤S132：对匹配矩阵中的元素进行筛选，获得匹配矩阵中小于第二阈值的元素集合；
第二阈值可以根据元件的实际情况自由设定；
步骤S133：在小于第二阈值的元素集合中，选取元素值最小的元素对应的待测图中的点，并在待测图中获取与模板图相同大小的图片作为待测图的子图，其中，子图的横向边缘与待测图的横向边缘平行，子图的纵向边缘与待测图的纵向边缘平行，子图的一个顶点为元素值最小的元素对应的待测图中的像素点；
在其中一个实施例中，获取模板图的灰度图和待测图的灰度图，并获得模板图的灰度图在待测图的灰度图中的匹配矩阵的步骤包括以下步骤：
分别获取模板图的灰度图和待测图的灰度图，使用归一化平方差匹配方法，以模板图的灰度图为卷积核，对待测图的灰度图进行卷积运算，获得模板图的灰度图在待测图的灰度图中的匹配矩阵。
通过上述方式获取的匹配矩阵能很好地反映待测图中的每个像素点与模板图的相似程度，匹配矩阵中小于第二阈值的元素越小，表明该元素对应的待测图中的像素点与模板图越匹配。
优选的，分别获取Imodel，Iobject的灰度图I’model，I’object，使用归一化平方差匹配方法，以I’model为卷积核，对I’object进行卷积运算，获得I’model在I’object中的匹配矩阵R，具体公式如下：
Figure PCTCN2016098226-appb-000001
其中，x、y的取值范围分别是，x∈[0,w_object–w_model]，y∈[0,h_object–h_model]。匹配矩阵R中每一个元素R(x,y)，都对应于I’object中的某个像素点的灰度值，也对应于Iobject中的相应的像素点。
在其中一个实施例中，对匹配矩阵中的元素进行筛选，获得匹配矩阵中小 于第一阈值的元素集合的步骤包括以下步骤：
对匹配矩阵R进行筛选，获得其中元素值小于第二阀值rthreshold的元素集合，若该元素集合为空集，则判定待测元件未出现在待测图中。
对于匹配矩阵中R中的元素进行筛选，选择小于第二阀值rthreshold的元素集合，可以避免对匹配矩阵中其他大于或者等于第二阈值的元素进行后续操作，这些元素就算进行后续操作也无法得到合适的结果，从而减少了工作量；而且对于元素集合为空集的情况，可直接判定待测元件未出现在待测图中，不必再进行后续操作。
在其中一个实施例中，获取子图的方式为：
在小于第二阈值的元素集合中，选取元素值最小的元素对应的待测图中的像素点，并在待测图中获取与模板图相同大小的图片作为待测图的子图，其中，子图的横向边缘与待测图的横向边缘平行，子图的纵向边缘与待测图的纵向边缘平行，子图的一个顶点为元素值最小的元素对应的待测图中的像素点。
通过上述方式，可以在待测图中获取一个与模板图相同大小，与模板图边缘位置平行的子图，只要在待测图中存在与模板图相同的待测元件，就能以子图的位置来确定待测元件在待测图的位置。
优选的，在小于第二阈值rthreshold的元素集合中，选取元素值最小的元素对应的待测图中的像素点作为最佳匹配点，该点坐标为(xbest,ybest)，对应的元素集合中的元素R(xbest,ybest)＝min(R(x,y))；
待测图Iobject中获取的子图Iobject_sub为
Iobject_sub(x,y)，x∈[xbest,xbest+wmodel)，y∈[ybest,ybest+hmodel)，
其中，xbest和ybest的取值范围分别为[0,wobject–wmodel]，[0,hobject–hmodel]，以最佳匹配点为顶点获取的子图的x最大值为wobject，y最大值为hobject，所以获取的子图始终在待测图的边界范围内。
在其中一个实施例中，计算模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度；若匹配度小于第一阈值，则判定子图所在的位置为待测元件在待测图中所在的位置的步骤如下：
根据模板图的颜色分布描述符CLDmodel和子图的颜色分布描述符 CLDobject_sub，计算两者之间的匹配度d，公式如下：
Figure PCTCN2016098226-appb-000002
如果d值小于第一阀值dthreshold，则子图Iobject_sub所在的位置就是待测元件在待测图中所在的位置。
在实际定位过程中，元件的位置不一定都是在预定的正确位置上，以最佳匹配点来确定子图，利用的是元件的图像位置信息；对比模板图的颜色描述分布符和子图的颜色描述分布符，利用的是元件的图像颜色信息，本发明方案结合了图像的形状信息和颜色信息，比普通定位算法准确性更高。
在其中一个实施例中，如图3所示，计算模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度的步骤之后包括以下步骤：
若匹配度大于或等于第一阈值，将顶点对应的小于第二阈值的元素集合中的元素清除，返回至在小于第二阈值的元素集合中，选取元素值最小的元素对应的待测图中的像素点的步骤，直至判定待测元件在待测图中所在的位置，或者元素集合成为空集为止。
在本实施例中，主要是对匹配度大于或等于第一阈值的对应的元素进行清除处理，从元素集合中重新选择；在实际定位过程中可能出现某元素的元素值最小，其匹配度不合适的情形，本实施例排除了此种情形。
优选的，若匹配度d大于或等于第一阈值dthreshold，则说明最佳匹配点(xbest,ybest)对应的子图在颜色空间中匹配度不够，就在元素集合中将最佳匹配点对应的小于第一阈值的元素集合中的元素清除，将清除相应元素后的元素集合，作为新的元素集合。
在新的元素集合中，选取元素值最小的元素对应的待测图中的点作为最佳匹配点，并根据上述方法选取新的子图，获取新的子图的颜色描述分布符，再计算模板图的颜色分布描述符和新的子图的颜色分布描述符之间的新匹配度，最后与第一阈值dthreshold进行比较，若新匹配度小于第一阈值dthreshold，则判定子图所在的位置就是待测元件在待测图中所在的位置；若大于或等于第一阈值dthreshold，则再次清除最佳匹配点对应的小于第二阈值的元素集合中的元素；重 复上述步骤，直至判定待测元件在待测图中所在的位置，或者元素集合成为空集为止；
另外，在选取元素值最小的元素时，若存在元素值相同的情况，则在元素值相同的若干个元素中随机选取一个。
在其中一个实施例中，若小于第二阈值的元素集合中的所有元素都被清除，元素集合成为空集时，则判定待测元件未出现在待测图中。
在本实施例中，出现元素集合中的所有元素都被清除的情形也能进行判断处理。
本发明提供了一种元件的定位方法，根据已知元件的图像，在待测图中定位元件的位置，定位准确，为对元件进行错，漏，反等检测提供了重要依据。避免了元件因为位移而不在检测区域中的问题，增强了元件检测结果的准确性,而且本发明结合了图像的形状信息和颜色信息，比普通定位算法准确性更高。
根据上述元件的定位方法，本发明还提供一种元件的定位系统，以下就本发明的元件的定位系统的实施例进行详细说明。
参见图4所示，为本发明的元件的定位系统的实施例。该实施例中的元件的定位系统包括第一获取单元210，第二获取单元220，第三获取单元230，第四获取单元240，计算单元250，判断单元260，其中：
第一获取单元210，用于获取待测元件的模板图和对待测元件实际拍摄的待测图；
第二获取单元220，用于获取模板图的颜色分布描述符；
第三获取单元230，用于在待测图中获取与模板图相同大小的图片作为待测图的子图；
第四获取单元240，用于获取子图的颜色分布描述符；
计算单元250，用于计算模板图的颜色分布描述符和子图的颜色分布描述符之间的匹配度；
判断单元260，用于在匹配度小于第一阈值时，判定子图所在的位置为待 测元件在待测图中所在的位置。
在其中一个实施例中，如图5和图6所示，第二获取单元220包括第一分图单元221、第一取值单元222和第一转换单元223；
第一分图单元221，用于将模板图均分成n×n个相同大小的单元图，n为正整数；
第一取值单元222，用于对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
第一转换单元223，用于将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的三个通道进行离散余弦变换，获得模板图的颜色分布描述符；
第四获取单元240包括第二分图单元241、第二取值单元242和第二转换单元243；
第二分图单元241，用于将子图均分成m×m个相同大小的单元图，m为正整数；
第二取值单元242，用于对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
第二转换单元243，用于将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的三个通道进行离散余弦变换，获得子图的颜色分布描述符。
在其中一个实施例中，如图7所示，第三获取单元230包括以下单元：
矩阵获取单元231，用于获取模板图的灰度图和待测图的灰度图，并获取模板图的灰度图在待测图的灰度图中的匹配矩阵，其中，匹配矩阵中的每个元素均对应于待测图中的像素点；
筛选单元232，用于对匹配矩阵中的元素进行筛选，获得匹配矩阵中小于第二阈值的元素集合；
子图获取单元233，用于在筛选单元获得匹配矩阵中小于第二阈值的元素集合后，在小于第二阈值的元素集合中，选取元素值最小的元素对应的待测图中的像素点，并在待测图中获取与模板图相同大小的图片作为待测图的子图， 其中，子图的横向边缘与待测图的横向边缘平行，子图的纵向边缘与待测图的纵向边缘平行，子图的一个顶点为元素值最小的元素对应的待测图中的像素点。
在其中一个实施例中，如图8所示，元件的定位系统还包括第二判断单元270，用于在获得的匹配矩阵中小于第二阈值的元素集合为空集时，判定待测元件未出现在待测图中。
在其中一个实施例中，如图9所示，元件的定位系统还包括清除单元280，
清除单元280用于在所述匹配度大于或等于第一阈值时，将顶点对应的小于第二阈值的元素集合中的元素清除；
子图获取单元233还用于在所述清除单元进行元素清除后，在小于第二阈值的元素集合中，选取元素值最小的元素对应的待测图中的像素点，并在待测图中获取与模板图相同大小的图片作为待测图的子图，其中，子图的横向边缘与待测图的横向边缘平行，子图的纵向边缘与待测图的纵向边缘平行，子图的一个顶点为元素值最小的元素对应的待测图中的像素点。
本实施例中是使部分单元循环运行，直至判定待测元件在待测图中所在的位置，或者元素集合成为空集为止。
在其中一个实施例中，如图10所示，第二判断单元270还用于在小于第二阈值的元素集合中的所有元素都被清除时，判定待测元件未出现在待测图中。
本发明的元件的定位系统与本发明的元件的定位方法一一对应，在上述元件的定位方法的实施例阐述的技术特征及其有益效果均适用于元件的定位系统的实施例中。
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)

  1. 一种元件的定位方法，其特征在于，包括以下步骤：
    获取待测元件的模板图和对所述待测元件实际拍摄的待测图；
    获取所述模板图的颜色分布描述符；
    在所述待测图中获取与所述模板图相同大小的图片作为所述待测图的子图；
    获取所述子图的颜色分布描述符；
    计算所述模板图的颜色分布描述符和所述子图的颜色分布描述符之间的匹配度；
    若所述匹配度小于第一阈值，则判定所述子图所在的位置为所述待测元件在所述待测图中所在的位置。
  2. 根据权利要求1所述的元件的定位方法，其特征在于，
    所述获取所述模板图的颜色分布描述符的步骤包括以下步骤：
    将所述模板图均分成n×n个相同大小的单元图，n为正整数，对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
    将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的Y、Cb、Cr三个通道进行离散余弦变换，获得所述模板图或所述子图的颜色分布描述符；
    所述获取所述子图的颜色分布描述符的步骤包括以下步骤：
    将所述子图均分成m×m相同大小的单元图，m为正整数，对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
    将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的Y、Cb、Cr三个通道进行离散余弦变换，获得所述子图的颜色分布描述符。
  3. 根据权利要求1所述的元件的定位方法，其特征在于，在所述待测图中获取与所述模板图相同大小的图片作为所述待测图的子图的步骤包括以下 步骤：
    获取所述模板图的灰度图和所述待测图的灰度图，并获取所述模板图的灰度图在所述待测图的灰度图中的匹配矩阵，其中，所述匹配矩阵中的每个元素均对应于所述待测图中的像素点；
    对所述匹配矩阵中的元素进行筛选，获得所述匹配矩阵中小于第二阈值的元素集合；
    在所述小于第二阈值的元素集合中，选取元素值最小的元素对应的所述待测图中的像素点，并在所述待测图中获取与所述模板图相同大小的图片作为所述待测图的子图，其中，所述子图的横向边缘与所述待测图的横向边缘平行，所述子图的纵向边缘与所述待测图的纵向边缘平行，所述子图的一个顶点为所述元素值最小的元素对应的所述待测图中的像素点。
  4. 根据权利要求3所述的元件的定位方法，其特征在于，所述对所述匹配矩阵中的元素进行筛选，获得所述匹配矩阵中小于第二阈值的元素集合的步骤包括以下步骤：
    对所述匹配矩阵中的元素进行筛选，若获得的所述匹配矩阵中小于第二阈值的元素集合为空集，则判定所述待测元件未出现在所述待测图中。
  5. 根据权利要求3所述的元件的定位方法，其特征在于，计算所述模板图的颜色分布描述符和所述子图的颜色分布描述符之间的匹配度的步骤之后包括以下步骤：
    若所述匹配度大于或等于第一阈值，将所述顶点对应的所述小于第二阈值的元素集合中的元素清除，返回至所述在所述小于第二阈值的元素集合中，选取元素值最小的元素对应的所述待测图中的像素点的步骤，直至判定所述待测元件在所述待测图中所在的位置，或者所述元素集合成为空集为止。
  6. 根据权利要求5所述的元件的定位方法，其特征在于，所述元素集合成为空集后包括以下步骤：
    若所述小于第二阈值的元素集合中的所有元素都被清除，所述元素集合成为空集，则判定所述待测元件未出现在所述待测图中。
  7. 一种元件的定位系统，其特征在于，包括以下单元：
    第一获取单元，用于获取待测元件的模板图和对所述待测元件实际拍摄的待测图；
    第二获取单元，用于获取所述模板图的颜色分布描述符；
    第三获取单元，用于在所述待测图中获取与所述模板图相同大小的图片作为所述待测图的子图；
    第四获取单元，用于获取所述子图的颜色分布描述符；
    计算单元，用于计算所述模板图的颜色分布描述符和所述子图的颜色分布描述符之间的匹配度；
    判断单元，用于在所述匹配度小于第一阈值时，判定所述子图所在的位置为所述待测元件在所述待测图中所在的位置。
  8. 根据权利要求7所述的元件的定位系统，其特征在于，
    所述第二获取单元包括第一分图单元、第一取值单元和第一转换单元；
    所述第一分图单元，用于将所述模板图均分成n×n个相同大小的单元图，n为正整数；
    所述第一取值单元，用于对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
    所述第一转换单元，用于将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的三个通道进行离散余弦变换，获得模板图的颜色分布描述符；
    所述第四获取单元包括第二分图单元、第二取值单元和第二转换单元；
    所述第二分图单元，用于将所述子图均分成m×m个相同大小的单元图，m为正整数；
    所述第二取值单元，用于对每一个单元图的RGB颜色空间的R、G、B三个通道分别取平均值，获得每一个单元图的平均颜色；
    所述第二转换单元，用于将每一个单元图的RGB颜色空间转变为YCbCr颜色空间，获得具有Y、Cb、Cr三个通道的转换图，对该转换图的三个通道进行离散余弦变换，获得所述子图的颜色分布描述符。
  9. 根据权利要求7所述的元件的定位系统，其特征在于，所述第三获取单元包括以下单元：
    矩阵获取单元，用于获取所述模板图的灰度图和所述待测图的灰度图，并获取所述模板图的灰度图在所述待测图的灰度图中的匹配矩阵，其中，所述匹配矩阵中的每个元素均对应于所述待测图中的像素点；
    筛选单元，用于对所述匹配矩阵中的元素进行筛选，获得所述匹配矩阵中小于第二阈值的元素集合；
    子图获取单元，用于在所述筛选单元获得所述匹配矩阵中小于第二阈值的元素集合后，在所述小于第二阈值的元素集合中，选取元素值最小的元素对应的所述待测图中的像素点，并在所述待测图中获取与所述模板图相同大小的图片作为所述待测图的子图，其中，所述子图的横向边缘与所述待测图的横向边缘平行，所述子图的纵向边缘与所述待测图的纵向边缘平行，所述子图的一个顶点为所述元素值最小的元素对应的所述待测图中的像素点。
  10. 根据权利要求9所述的元件的定位系统，其特征在于，所述元件的定位系统还包括清除单元；
    所述清除单元用于在所述匹配度大于或等于第一阈值时，将所述顶点对应的所述小于第二阈值的元素集合中的元素清除；
    所述子图获取单元还用于在所述清除单元进行元素清除后，在所述小于第二阈值的元素集合中，选取元素值最小的元素对应的所述待测图中的像素点，并在所述待测图中获取与所述模板图相同大小的图片作为所述待测图的子图，其中，所述子图的横向边缘与所述待测图的横向边缘平行，所述子图的纵向边缘与所述待测图的纵向边缘平行，所述子图的一个顶点为所述元素值最小的元素对应的所述待测图中的像素点。
PCT/CN2016/098226 2015-11-12 2016-09-06 元件的定位方法和系统 WO2017080295A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510770846.8 2015-11-12
CN201510770846.8A CN105389818B (zh) 2015-11-12 2015-11-12 元件的定位方法和系统

Publications (1)

Publication Number Publication Date
WO2017080295A1 true WO2017080295A1 (zh) 2017-05-18

Family

ID=55422067

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/098226 WO2017080295A1 (zh) 2015-11-12 2016-09-06 元件的定位方法和系统

Country Status (2)

Country Link
CN (1) CN105389818B (zh)
WO (1) WO2017080295A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105389818B (zh) * 2015-11-12 2018-11-27 广州视源电子科技股份有限公司 元件的定位方法和系统
CN106485710A (zh) * 2016-10-18 2017-03-08 广州视源电子科技股份有限公司 一种元件错件检测方法和装置
CN106485284B (zh) * 2016-10-19 2019-05-14 哈尔滨工业大学 一种基于模板匹配的元件定位方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007309808A (ja) * 2006-05-19 2007-11-29 Juki Corp 対象物の位置検出方法及び装置
CN102486829A (zh) * 2010-12-01 2012-06-06 鸿富锦精密工业（深圳）有限公司 图像分析系统及方法
CN102914549A (zh) * 2012-09-10 2013-02-06 中国航天科技集团公司第五研究院第五一三研究所 针对星载表露型pcb焊点质量的光学图像匹配检测方法
CN103559499A (zh) * 2013-10-09 2014-02-05 华南理工大学 一种rgb矢量匹配快速识别系统及方法
CN104504375A (zh) * 2014-12-18 2015-04-08 广州视源电子科技股份有限公司 一种pcb元件的识别方法及装置
CN105389818A (zh) * 2015-11-12 2016-03-09 广州视源电子科技股份有限公司 元件的定位方法和系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007309808A (ja) * 2006-05-19 2007-11-29 Juki Corp 対象物の位置検出方法及び装置
CN102486829A (zh) * 2010-12-01 2012-06-06 鸿富锦精密工业（深圳）有限公司 图像分析系统及方法
CN102914549A (zh) * 2012-09-10 2013-02-06 中国航天科技集团公司第五研究院第五一三研究所 针对星载表露型pcb焊点质量的光学图像匹配检测方法
CN103559499A (zh) * 2013-10-09 2014-02-05 华南理工大学 一种rgb矢量匹配快速识别系统及方法
CN104504375A (zh) * 2014-12-18 2015-04-08 广州视源电子科技股份有限公司 一种pcb元件的识别方法及装置
CN105389818A (zh) * 2015-11-12 2016-03-09 广州视源电子科技股份有限公司 元件的定位方法和系统

Also Published As

Publication number Publication date
CN105389818B (zh) 2018-11-27
CN105389818A (zh) 2016-03-09

Similar Documents

Publication Publication Date Title
WO2017177717A1 (zh) 基于颜色和梯度的元件定位方法和系统
EP3989542A1 (en) Method and apparatus for evaluating image acquisition accuracy, and electronic device and storage medium
CN108520514B (zh) 基于计算机视觉的印刷电路板电子元器一致性检测方法
AU2019336484B2 (en) Digital assessment of chemical dip tests
WO2018068417A1 (zh) 元件缺陷检测方法和系统
WO2017080295A1 (zh) 元件的定位方法和系统
CN107239742A (zh) 一种仪表指针刻度值计算方法
CN107271445B (zh) 一种缺陷检测方法及装置
KR101359280B1 (ko) 패턴 매칭 방법, 패턴 매칭 프로그램을 기록한 컴퓨터로 판독가능한 기록 매체, 전자 계산기, 전자 디바이스 검사 장치
WO2017071406A1 (zh) 金针类元件的引脚检测方法和系统
JP5660861B2 (ja) 基板上の異物検査方法および異物検査装置
Flesia et al. Sub-pixel straight lines detection for measuring through machine vision
CN111768348A (zh) 一种缺陷检测方法、设备及计算机可读存储介质
CN112782179A (zh) 一种产品反光表面缺陷检测方法及系统
WO2017181722A1 (zh) 元件漏检方法和系统
TW201326735A (zh) 寬度量測方法及系統
CN117058411B (zh) 电池棱边外形瑕疵识别方法、装置、介质和设备
CN114187363A (zh) 一种获取径向畸变参数值的方法、装置以及移动终端
JP2005345290A (ja) 筋状欠陥検出方法及び装置
CN115375608A (zh) 检测方法及装置、检测设备和存储介质
CN113176275A (zh) 一种用于显示面板复检的方法、装置及系统
WO2018094823A1 (zh) 显示屏均匀性测试方法及系统
CN116879708A (zh) 一种利用红外云图识别技术印刷电路板故障的检测方法
CN109448012A (zh) 一种图像边缘检测方法及装置
CN109215068B (zh) 图像放大率测量方法及装置

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: 16863473

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: 16863473

Country of ref document: EP

Kind code of ref document: A1