WO2018068417A1 - 元件缺陷检测方法和系统 - Google Patents

元件缺陷检测方法和系统 Download PDF

Info

Publication number
WO2018068417A1
WO2018068417A1 PCT/CN2016/113148 CN2016113148W WO2018068417A1 WO 2018068417 A1 WO2018068417 A1 WO 2018068417A1 CN 2016113148 W CN2016113148 W CN 2016113148W WO 2018068417 A1 WO2018068417 A1 WO 2018068417A1
Authority
WO
WIPO (PCT)
Prior art keywords
contour
image
pixel
component
similarity
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.)
Ceased
Application number
PCT/CN2016/113148
Other languages
English (en)
French (fr)
Inventor
李红匣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Shiyuan Electronics Thecnology Co Ltd
Original Assignee
Guangzhou Shiyuan Electronics Thecnology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangzhou Shiyuan Electronics Thecnology Co Ltd filed Critical Guangzhou Shiyuan Electronics Thecnology Co Ltd
Publication of WO2018068417A1 publication Critical patent/WO2018068417A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30108Industrial image inspection
    • G06T2207/30141Printed circuit board [PCB]

Definitions

  • the invention relates to the field of automatic optical detection technology, and in particular to a component defect detection method and system.
  • AOI Automatic Optic Inspection
  • common defect detection includes missing parts detection, wrong part detection, reverse part detection, multi-piece detection, and the like.
  • the fault detection refers to extracting the features of the component to be detected and comparing with the template to determine whether the components inserted into the circuit board are correct.
  • the wrong component detection of components is mainly detected by manual, but this detection method is inefficient, and the detection result is easy to make mistakes, and the detection accuracy is low.
  • the existing defect detection mode has low detection efficiency and low detection accuracy.
  • a component defect detecting method includes the following steps:
  • a component defect detecting system includes:
  • An acquiring module configured to acquire an original image of the device to be tested mounted on the circuit board, perform edge detection on the original image, and acquire a contour image of the device to be tested;
  • a calculation module configured to extract a contour feature of the contour image, compare the contour feature with a template contour feature of the pre-stored template contour image, and calculate a similarity between the contour feature and the template contour feature;
  • the determining module is configured to determine that the component is installed incorrectly if the similarity is less than a preset similarity threshold.
  • the above component defect detecting method and system by acquiring a contour image of the component to be tested, and calculating a similarity between the contour feature of the contour image and the template contour feature of the pre-stored template contour image, when the similarity is less than a preset similarity threshold
  • the component is determined to be installed incorrectly, and the component defect automatic detection is realized, and the detection efficiency and the accuracy are high.
  • FIG. 1 is a flow chart of a component defect detecting method of an embodiment
  • FIG. 2 is a schematic view showing the structure of a component defect detecting system of one embodiment.
  • the component defect detecting method may include the following steps:
  • the original image of the device to be tested may be first located from the image of the entire circuit board, and the image of the feature region of the device to be tested is located from the original image.
  • the original images of the components to be tested can be separately obtained, and then the edge detection of each original image is performed separately.
  • the original images corresponding to the respective components to be tested may be sequentially stored according to the positions of the respective components to be tested on the circuit board, so as to facilitate the execution of the subsequent detection operations.
  • each of the original images may also be sequentially numbered to facilitate execution of subsequent detection operations.
  • the contour image of the device to be tested can be obtained in the following manner:
  • Step 1 Calculate pixel values of respective pixels on the original image
  • Step 2 respectively compare the pixel values of the respective pixel points with a preset pixel threshold
  • Step 3 If the pixel value of the pixel is greater than the pixel threshold, and the pixel has an adjacent pixel smaller than the pixel threshold, the pixel is set as an edge pixel;
  • Step 4 Acquire a contour image of the device under test according to each edge pixel.
  • the original image may be subjected to noise reduction processing before comparing the pixel value with a preset pixel threshold.
  • the original image is subjected to gradation processing to obtain a grayscale image, and the grayscale image is binarized according to a preset pixel threshold.
  • the original image may be enlarged before being compared.
  • the contour features may include, but are not limited to, a center position of the contour point, a curvature of the contour, and/or an area of the area enclosed by the contour point.
  • the manner of calculating the similarity between the contour feature and the template contour feature may include three types:
  • a first center position of each contour point of the contour image and a second center position of each contour point of the template contour image may be calculated, The first central location is compared to the second central location.
  • the contour feature is the curvature of the contour
  • the first curvature of the contour formed by each contour point of the contour image and the second curvature of the contour formed by each contour point of the template contour image may be calculated. Comparing the first curvature with the second curvature.
  • a first area of the contour area surrounded by each contour point of the contour image and each contour point of the template contour image may be calculated.
  • the second area of the enclosed contour area is compared to the first area and the second area.
  • step S2 If the degree of similarity between the contour feature and the template contour feature is calculated in step S2, in this step, if the distance between the first central location and the second central location is greater than a preset The distance threshold can be used to determine that the component is installed incorrectly; otherwise, the component can be determined to be properly installed.
  • step S2 If the difference between the contour feature and the template contour feature is calculated by using the second method in step S2, in this step, if the difference between the first curvature and the second curvature is greater than a preset curvature threshold, Determining that the component is installed incorrectly Wrong; otherwise, it can be determined that the component is installed correctly.
  • step S2 If the similarity between the contour feature and the template contour feature is calculated by using the third method in step S2, in this step, if the difference between the first area and the second area is greater than a preset area threshold, It is determined that the component is installed incorrectly; otherwise, it can be determined that the component is properly installed.
  • the above distance threshold, curvature threshold and area threshold may be set according to actual needs (for example, accuracy requirement, calculation amount requirement or calculation rate requirement, etc.), or may be calculated by using a correlation algorithm.
  • the above component defect detecting method realizes automatic detection of component defects, and has high detection efficiency and high accuracy. Especially when the colors of the components are similar, or the character characteristics of the components are similar, it is possible to effectively detect whether the components are installed incorrectly. Moreover, the above component defect detecting method can select the most suitable similarity calculating method according to actual conditions, and further improve the detection accuracy.
  • the component defect detecting system may include:
  • the acquiring module 10 is configured to acquire an original image of the device to be tested mounted on the circuit board, perform edge detection on the original image, and acquire a contour image of the device to be tested;
  • the original image of the device to be tested may be first located from the image of the entire circuit board, and the image of the feature region of the device to be tested is located from the original image.
  • the original images of the components to be tested can be separately obtained, and then the edge detection of each original image is performed separately.
  • the original images corresponding to the respective components to be tested may be sequentially stored according to the positions of the respective components to be tested on the circuit board, so as to facilitate the execution of the subsequent detection operations.
  • each of the original images may also be sequentially numbered to facilitate execution of subsequent detection operations.
  • the obtaining module 10 can include:
  • a calculating unit configured to calculate a pixel value of each pixel on the original image
  • a comparing unit configured to respectively compare pixel values of the respective pixel points with a preset pixel threshold
  • a setting unit configured to: if the pixel value of the pixel point is greater than the pixel threshold, and the pixel point has an adjacent pixel point smaller than the pixel threshold, set the pixel point as an edge pixel point;
  • an acquiring unit configured to acquire a contour image of the device to be tested according to each edge pixel.
  • the original image may be subjected to noise reduction processing before comparing the pixel value with a preset pixel threshold.
  • the original image is subjected to gradation processing to obtain a grayscale image, and the grayscale image is binarized according to a preset pixel threshold.
  • a calculation module 20 configured to extract a contour feature of the contour image, compare the contour feature with a template contour feature of the pre-stored template contour image, and calculate a similarity between the contour feature and the template contour feature;
  • the contour features may include, but are not limited to, a center position of the contour point, a curvature of the contour, and/or an area of the area enclosed by the contour point.
  • the manner of calculating the similarity between the contour feature and the template contour feature may include three types:
  • a first center position of each contour point of the contour image and a second center position of each contour point of the template contour image may be calculated, The first central location is compared to the second central location.
  • the contour feature is the curvature of the contour
  • the first curvature of the contour formed by each contour point of the contour image and the second curvature of the contour formed by each contour point of the template contour image may be calculated. Comparing the first curvature with the second curvature.
  • a first area of the contour area surrounded by each contour point of the contour image and each contour point of the template contour image may be calculated.
  • the second area of the enclosed contour area is compared to the first area and the second area.
  • the determining module 30 is configured to determine that the component is installed incorrectly if the similarity is less than a preset similarity threshold.
  • calculation module 20 uses the first method to calculate the similarity between the contour feature and the template contour feature, in this step, if the distance between the first central location and the second central location is greater than the pre- The distance threshold is set to determine that the component is installed incorrectly; otherwise, the component can be determined to be properly installed.
  • calculation module 20 uses the second method to calculate the similarity between the contour feature and the template contour feature, in this step, if the difference between the first curvature and the second curvature is greater than a preset curvature threshold Determining the component Error loading; otherwise, it can be determined that the components are installed correctly.
  • the similarity between the contour feature and the template contour feature is calculated by using the third method in the calculation module 20, in this step, if the difference between the first area and the second area is greater than a preset area threshold It is determined that the component is installed incorrectly; otherwise, it can be determined that the component is correctly installed.
  • the above distance threshold, curvature threshold and area threshold may be set according to actual needs (for example, accuracy requirement, calculation amount requirement or calculation rate requirement, etc.), or may be calculated by using a correlation algorithm.
  • the above component defect detecting system realizes automatic detection of component defects, and has high detection efficiency and high accuracy. Especially when the colors of the components are similar, or the character characteristics of the components are similar, it is possible to effectively detect whether the components are installed incorrectly. Moreover, the component defect detecting system can select the most suitable similarity calculation method according to actual conditions, thereby further improving the detection accuracy.
  • the component defect detecting system of the present invention has a one-to-one correspondence with the component defect detecting method of the present invention, and the technical features and the advantageous effects thereof described in the embodiments of the component defect detecting method are applicable to the embodiment of the component defect detecting system, and hereby declare .

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Image Processing (AREA)

Abstract

一种元件缺陷检测方法和系统,方法包括以下步骤:获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像(S1);提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度(S2);若所述相似度小于预设的相似度阈值,判定所述元件安装错误(S3)。

Description

元件缺陷检测方法和系统 技术领域
本发明涉及自动光学检测技术领域,特别是涉及一种元件缺陷检测方法和系统。
背景技术
AOI(Automatic Optic Inspection,自动光学检测),是利用光学原理对电路板焊接生产中出现的常见缺陷进行检测的设备。对于插件的电路板来说,常见的缺陷检测包括漏件检测、错件检测、反件检测、多件检测等。以元件错件检测为例,错件检测是指提取待检测元件的特征,并与模板进行比较,从而判断插入电路板的元件是否正确。
目前,元件的错件检测主要由人工进行检测,但是,这种检测方式效率较低,而且,检测结果容易出错,检测正确率较低。
综上所述,现有的缺陷检测方式检测效率和检测正确率较低。
发明内容
基于此,有必要针对检测效率和正确率低的问题,提供一种元件缺陷检测方法和系统。
一种元件缺陷检测方法,包括以下步骤:
获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像;
提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度;
若所述相似度小于预设的相似度阈值,判定所述元件安装错误。
一种元件缺陷检测系统,包括:
获取模块,用于获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像;
计算模块,用于提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度;
判断模块,用于若所述相似度小于预设的相似度阈值,判定所述元件安装错误。
上述元件缺陷检测方法和系统,通过获取待测元件的轮廓图像,并计算该轮廓图像的轮廓特征与预存的模板轮廓图像的模板轮廓特征的相似度,当相似度小于预设的相似度阈值时,判定所述元件安装错误,实现了元件缺陷自动检测,检测效率和准确率高。
附图说明
图1为一个实施例的元件缺陷检测方法流程图;
图2为一个实施例的元件缺陷检测系统的结构示意图。
具体实施方式
下面结合附图对本发明的技术方案进行说明。
图1为一个实施例的元件缺陷检测方法流程图。如图1所示,所述元件缺陷检测方法可包括以下步骤:
S1,获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像;
在检测前,可以首先从整个电路板的图像中定位出所述待测元件的原始图像,再从所述原始图像中定位出所述待测元件的特征区域图像。当一块电路板上有多个待测元件都需要进行错件检测时,可以分别获取各个待测元件的原始图像,再分别对各个原始图像进行边缘检测。可以根据各个待测元件在电路板上的位置对各个待测元件对应的原始图像进行顺序存储,以便于后续检测操作的执行。在其中一个实施例中,还可以为各个原始图像顺序编号,以便于后续检测操作的执行。
可具体通过以下方式获取所述待测元件的轮廓图像:
步骤1:计算所述原始图像上各个像素点的像素值;
步骤2:分别将各个像素点的像素值与预设的像素阈值进行比较;
步骤3:若所述像素点的像素值大于所述像素阈值,且所述像素点存在小于所述像素阈值的相邻像素点,将所述像素点设为边缘像素点;
步骤4:根据各个边缘像素点获取所述待测元件的轮廓图像。
为了消除电路板上的污点以及背景颜色和图案等因素对检测结果的影响,在将所述像素值与预设的像素阈值进行比较之前,可以对所述原始图像进行降噪处理。具体地,可以 对所述原始图像进行灰度处理,得到灰度图像,并根据预设的像素阈值对所述灰度图像进行二值化处理。
S2,提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度;
在本步骤中,为了防止元件尺寸太小,不利于后续操作,在进行比较之前,还可以对原始图像进行放大处理。所述放大处理是指尺寸的放大,即将图像的长宽分别放大到原来的n倍,n可以根据实际的需求设置,一般情况n=2即可。
在实际应用中,所述轮廓特征可包括但不限于轮廓点的中心位置、轮廓的曲率和/或轮廓点所围成的区域的面积。对应地,计算所述轮廓特征与所述模板轮廓特征之间的相似度的方式可包括三种:
方式一,当所述轮廓特征为轮廓点的中心位置时,可计算所述轮廓图像的各个轮廓点的第一中心位置和所述模板轮廓图像的各个轮廓点的第二中心位置,将所述第一中心位置和所述第二中心位置进行比较。该方案适用于元件在图像中不易发生偏移、且形状不规则的情况。
方式二,当所述轮廓特征为轮廓的曲率时,可以计算所述轮廓图像的各个轮廓点所构成的轮廓的第一曲率和所述模板轮廓图像的各个轮廓点所构成的轮廓的第二曲率,将所述第一曲率和所述第二曲率进行比较。该方案适用于待测元件与模板元件的轮廓形状不同的情况。
方式三,当所述轮廓特征为轮廓点所围成的区域的面积时,可以计算所述轮廓图像的各个轮廓点所围成的轮廓区域的第一面积和所述模板轮廓图像的各个轮廓点所围成的轮廓区域的第二面积,将所述第一面积和所述第二面积进行比较。该方案适用于待测元件与模板元件的形状相似、但元件顶部大小不同的情况。
S3,若所述相似度小于预设的相似度阈值,判定所述元件安装错误。
若步骤S2中采用方式一计算所述轮廓特征与所述模板轮廓特征之间的相似度,在本步骤中,若所述第一中心位置和所述第二中心位置之间的距离大于预设的距离阈值,可判定所述元件安装错误;否则,可判定所述元件安装正确。
若步骤S2中采用方式二计算所述轮廓特征与所述模板轮廓特征之间的相似度,在本步骤中,若所述第一曲率和所述第二曲率之差大于预设的曲率阈值,判定所述元件安装错 误;否则,可判定所述元件安装正确。
若步骤S2中采用方式三计算所述轮廓特征与所述模板轮廓特征之间的相似度,在本步骤中,若所述第一面积和所述第二面积之差大于预设的面积阈值,判定所述元件安装错误;否则,可判定所述元件安装正确。
上述距离阈值、曲率阈值和面积阈值可以根据实际需要(例如,精确度要求、计算量要求或计算速率要求等)自行设定,也可以采用相关算法计算。
上述元件缺陷检测方法,实现了元件缺陷自动检测,检测效率和准确率高。尤其是在元件的颜色相近,或元件外表的字符特征也相似时,能够有效检测元件是否安装错误。并且,上述元件缺陷检测方法能够根据实际情况选择最适合的相似度计算方式,进一步提高了检测准确性。
图2为一个实施例的元件缺陷检测系统的结构示意图。如图2所示,所述元件缺陷检测系统可包括:
获取模块10,用于获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像;
在检测前,可以首先从整个电路板的图像中定位出所述待测元件的原始图像,再从所述原始图像中定位出所述待测元件的特征区域图像。当一块电路板上有多个待测元件都需要进行错件检测时,可以分别获取各个待测元件的原始图像,再分别对各个原始图像进行边缘检测。可以根据各个待测元件在电路板上的位置对各个待测元件对应的原始图像进行顺序存储,以便于后续检测操作的执行。在其中一个实施例中,还可以为各个原始图像顺序编号,以便于后续检测操作的执行。
所述获取模块10可包括:
计算单元,用于计算所述原始图像上各个像素点的像素值;
比较单元,用于分别将各个像素点的像素值与预设的像素阈值进行比较;
设置单元,用于若所述像素点的像素值大于所述像素阈值,且所述像素点存在小于所述像素阈值的相邻像素点,将所述像素点设为边缘像素点;
获取单元,用于根据各个边缘像素点获取所述待测元件的轮廓图像。
为了消除电路板上的污点以及背景颜色和图案等因素对检测结果的影响,在将所述像素值与预设的像素阈值进行比较之前,可以对所述原始图像进行降噪处理。具体地,可以 对所述原始图像进行灰度处理,得到灰度图像,并根据预设的像素阈值对所述灰度图像进行二值化处理。
计算模块20,用于提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度;
为了防止元件尺寸太小,不利于后续操作,在进行比较之前,还可以对原始图像进行放大处理。所述放大处理是指尺寸的放大,即将图像的长宽分别放大到原来的n倍,n可以根据实际的需求设置,一般情况n=2即可。
在实际应用中,所述轮廓特征可包括但不限于轮廓点的中心位置、轮廓的曲率和/或轮廓点所围成的区域的面积。对应地,计算所述轮廓特征与所述模板轮廓特征之间的相似度的方式可包括三种:
方式一,当所述轮廓特征为轮廓点的中心位置时,可计算所述轮廓图像的各个轮廓点的第一中心位置和所述模板轮廓图像的各个轮廓点的第二中心位置,将所述第一中心位置和所述第二中心位置进行比较。该方案适用于元件在图像中不易发生偏移、且形状不规则的情况。
方式二,当所述轮廓特征为轮廓的曲率时,可以计算所述轮廓图像的各个轮廓点所构成的轮廓的第一曲率和所述模板轮廓图像的各个轮廓点所构成的轮廓的第二曲率,将所述第一曲率和所述第二曲率进行比较。该方案适用于待测元件与模板元件的轮廓形状不同的情况。
方式三,当所述轮廓特征为轮廓点所围成的区域的面积时,可以计算所述轮廓图像的各个轮廓点所围成的轮廓区域的第一面积和所述模板轮廓图像的各个轮廓点所围成的轮廓区域的第二面积,将所述第一面积和所述第二面积进行比较。该方案适用于待测元件与模板元件的形状相似、但元件顶部大小不同的情况。
判断模块30,用于若所述相似度小于预设的相似度阈值,判定所述元件安装错误。
若计算模块20中采用方式一计算所述轮廓特征与所述模板轮廓特征之间的相似度,在本步骤中,若所述第一中心位置和所述第二中心位置之间的距离大于预设的距离阈值,可判定所述元件安装错误;否则,可判定所述元件安装正确。
若计算模块20中采用方式二计算所述轮廓特征与所述模板轮廓特征之间的相似度,在本步骤中,若所述第一曲率和所述第二曲率之差大于预设的曲率阈值,判定所述元件安 装错误;否则,可判定所述元件安装正确。
若计算模块20中采用方式三计算所述轮廓特征与所述模板轮廓特征之间的相似度,在本步骤中,若所述第一面积和所述第二面积之差大于预设的面积阈值,判定所述元件安装错误;否则,可判定所述元件安装正确。
上述距离阈值、曲率阈值和面积阈值可以根据实际需要(例如,精确度要求、计算量要求或计算速率要求等)自行设定,也可以采用相关算法计算。
上述元件缺陷检测系统,实现了元件缺陷自动检测,检测效率和准确率高。尤其是在元件的颜色相近,或元件外表的字符特征也相似时,能够有效检测元件是否安装错误。并且,上述元件缺陷检测系统能够根据实际情况选择最适合的相似度计算方式,进一步提高了检测准确性。
本发明的元件缺陷检测系统与本发明的元件缺陷检测方法一一对应,在上述元件缺陷检测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缺陷检测系统的实施例中,特此声明。
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. 一种元件缺陷检测方法,其特征在于,包括以下步骤:
    获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像;
    提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度;
    若所述相似度小于预设的相似度阈值,判定所述元件安装错误。
  2. 根据权利要求1所述的元件缺陷检测方法,其特征在于,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像的步骤包括:
    计算所述原始图像上各个像素点的像素值;
    分别将各个像素点的像素值与预设的像素阈值进行比较;
    若所述像素点的像素值大于所述像素阈值,且所述像素点存在小于所述像素阈值的相邻像素点,将所述像素点设为边缘像素点;
    根据各个边缘像素点获取所述待测元件的轮廓图像。
  3. 根据权利要求2所述的元件缺陷检测方法,其特征在于,在将所述像素值与预设的像素阈值进行比较之前,还包括以下步骤:
    对所述原始图像进行灰度处理,得到灰度图像;
    对所述灰度图像进行二值化处理。
  4. 根据权利要求1所述的元件缺陷检测方法,其特征在于,所述轮廓特征包括轮廓点的中心位置、轮廓的曲率和/或轮廓点所围成的区域的面积。
  5. 根据权利要求4所述的元件缺陷检测方法,其特征在于,计算所述轮廓图像的轮廓特征与所述模板轮廓图像的轮廓特征之间的相似度的步骤包括:
    计算所述轮廓图像的各个轮廓点的第一中心位置和所述模板轮廓图像的各个轮廓点的第二中心位置;
    将所述第一中心位置和所述第二中心位置进行比较;
    若所述相似度小于预设的相似度阈值,判定所述元件安装错误的步骤包括:
    若所述第一中心位置和所述第二中心位置之间的距离大于预设的距离阈值,判定所述元件安装错误。
  6. 根据权利要求4所述的元件缺陷检测方法,其特征在于,计算所述轮廓图像的轮廓特征与所述模板轮廓图像的轮廓特征之间的相似度的步骤还包括:
    计算所述轮廓图像的各个轮廓点所构成的轮廓的第一曲率和所述模板轮廓图像的各个轮廓点所构成的轮廓的第二曲率;
    将所述第一曲率和所述第二曲率进行比较;
    若所述相似度小于预设的相似度阈值,判定所述元件安装错误的步骤包括:
    若所述第一曲率和所述第二曲率之差大于预设的曲率阈值,判定所述元件安装错误。
  7. 根据权利要求4所述的元件缺陷检测方法,其特征在于,计算所述轮廓图像的轮廓特征与所述模板轮廓图像的轮廓特征之间的相似度的步骤还包括:
    计算所述轮廓图像的各个轮廓点所围成的轮廓区域的第一面积和所述模板轮廓图像的各个轮廓点所围成的轮廓区域的第二面积;
    将所述第一面积和所述第二面积进行比较;
    若所述相似度小于预设的相似度阈值,判定所述元件安装错误的步骤包括:
    若所述第一面积和所述第二面积之差大于预设的面积阈值,判定所述元件安装错误。
  8. 一种元件缺陷检测系统,其特征在于,包括:
    获取模块,用于获取安装在电路板上的待测元件的原始图像,对所述原始图像进行边缘检测,获取所述待测元件的轮廓图像;
    计算模块,用于提取所述轮廓图像的轮廓特征,将所述轮廓特征与预存的模板轮廓图像的模板轮廓特征进行比较,计算所述轮廓特征与所述模板轮廓特征之间的相似度;
    判断模块,用于若所述相似度小于预设的相似度阈值,判定所述元件安装错误。
  9. 根据权利要求8所述的元件缺陷检测系统,其特征在于,所述获取模块包括:
    计算单元,用于计算所述原始图像上各个像素点的像素值;
    比较单元,用于分别将各个像素点的像素值与预设的像素阈值进行比较;
    设置单元,用于若所述像素点的像素值大于所述像素阈值,且所述像素点存在小于所述像素阈值的相邻像素点,将所述像素点设为边缘像素点;
    获取单元,用于根据各个边缘像素点获取所述待测元件的轮廓图像。
  10. 根据权利要求8所述的元件缺陷检测系统,其特征在于,所述轮廓特征包括轮廓点的中心位置、轮廓的曲率和/或轮廓点所围成的区域的面积。
PCT/CN2016/113148 2016-10-11 2016-12-29 元件缺陷检测方法和系统 Ceased WO2018068417A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610887937.4 2016-10-11
CN201610887937.4A CN106504231A (zh) 2016-10-11 2016-10-11 元件缺陷检测方法和系统

Publications (1)

Publication Number Publication Date
WO2018068417A1 true WO2018068417A1 (zh) 2018-04-19

Family

ID=58293907

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/113148 Ceased WO2018068417A1 (zh) 2016-10-11 2016-12-29 元件缺陷检测方法和系统

Country Status (2)

Country Link
CN (1) CN106504231A (zh)
WO (1) WO2018068417A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4350638A4 (en) * 2021-05-24 2025-05-21 ZTE Corporation DETECTION METHOD AND SYSTEM FOR BASE STATION POWER LINE, AND RELATED APPARATUS
CN121329969A (zh) * 2025-12-15 2026-01-13 宁波星鼎电器有限公司 基于视觉的配件外观缺陷检测方法及系统

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107292901B (zh) * 2017-07-24 2020-06-02 北京小米移动软件有限公司 边缘检测方法及装置
CN107886495B (zh) * 2017-09-30 2020-04-14 北京得华机器人技术研究院有限公司 一种基于相似度匹配的汽车零配件缺陷识别方法
CN107680091B (zh) * 2017-10-12 2020-11-17 成都思晗科技股份有限公司 一种输电线路的防震锤脱落缺陷检测方法
CN109447961A (zh) * 2018-10-19 2019-03-08 闻泰通讯股份有限公司 内观检测方法及其装置
CN110136123A (zh) * 2019-05-17 2019-08-16 无锡睿勤科技有限公司 物品检测方法、移动终端及计算机可读存储介质
CN110596120A (zh) * 2019-09-06 2019-12-20 深圳新视智科技术有限公司 玻璃边界缺陷检测方法、装置、终端及存储介质
CN113554579A (zh) * 2020-04-07 2021-10-26 东华大学 皮革的缺陷检测方法
CN112183618B (zh) * 2020-09-25 2024-03-26 杭州三坛医疗科技有限公司 相似度确定方法和相似度确定装置
CN114636669B (zh) * 2020-12-15 2025-10-17 广东美的白色家电技术创新中心有限公司 电路板检测方法、检测系统及具有存储功能的装置
CN114723648A (zh) * 2020-12-18 2022-07-08 广东美的白色家电技术创新中心有限公司 浮高检测方法、检测系统及具有存储功能的装置
CN113643296B (zh) * 2021-10-18 2022-02-08 季华实验室 电感元件装配质量检测方法、装置、电子设备及存储介质
CN115184381B (zh) * 2022-08-05 2025-10-03 华研天创(珠海)科技有限公司 基于图像识别的pcb板检测装置及图像识别方法
CN115587966B (zh) * 2022-08-31 2025-12-30 山东省科学院自动化研究所 一种在光照不均匀条件下零件有无缺漏的检测方法及系统
CN115456986A (zh) * 2022-09-09 2022-12-09 湖北三江航天红林探控有限公司 基于图像处理的印制板质量检测系统及方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101055620A (zh) * 2006-04-12 2007-10-17 富士通株式会社 形状比较装置和方法
CN102831381A (zh) * 2011-06-15 2012-12-19 鸿富锦精密工业(深圳)有限公司 影像差异比较系统及方法
CN105957059A (zh) * 2016-04-20 2016-09-21 广州视源电子科技股份有限公司 电子元件漏件检测方法和系统
CN106530275A (zh) * 2016-10-11 2017-03-22 广州视源电子科技股份有限公司 元件错件检测方法和系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6868175B1 (en) * 1999-08-26 2005-03-15 Nanogeometry Research Pattern inspection apparatus, pattern inspection method, and recording medium
CN102721364B (zh) * 2011-03-30 2015-12-02 比亚迪股份有限公司 一种工件的定位方法及其装置
JP6557943B2 (ja) * 2014-01-15 2019-08-14 オムロン株式会社 画像照合装置、画像センサ、処理システム、画像照合方法
JP6496159B2 (ja) * 2015-02-23 2019-04-03 株式会社Screenホールディングス パターン検査装置およびパターン検査方法
CN105335973B (zh) * 2015-10-20 2018-08-03 上海交通大学 运用于带钢加工生产线的视觉处理方法
CN105405142B (zh) * 2015-11-12 2019-04-05 冯平 一种玻璃面板的边缺陷检测方法及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101055620A (zh) * 2006-04-12 2007-10-17 富士通株式会社 形状比较装置和方法
CN102831381A (zh) * 2011-06-15 2012-12-19 鸿富锦精密工业(深圳)有限公司 影像差异比较系统及方法
CN105957059A (zh) * 2016-04-20 2016-09-21 广州视源电子科技股份有限公司 电子元件漏件检测方法和系统
CN106530275A (zh) * 2016-10-11 2017-03-22 广州视源电子科技股份有限公司 元件错件检测方法和系统

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4350638A4 (en) * 2021-05-24 2025-05-21 ZTE Corporation DETECTION METHOD AND SYSTEM FOR BASE STATION POWER LINE, AND RELATED APPARATUS
CN121329969A (zh) * 2025-12-15 2026-01-13 宁波星鼎电器有限公司 基于视觉的配件外观缺陷检测方法及系统

Also Published As

Publication number Publication date
CN106504231A (zh) 2017-03-15

Similar Documents

Publication Publication Date Title
WO2018068417A1 (zh) 元件缺陷检测方法和系统
WO2018068415A1 (zh) 元件错件检测方法和系统
US20210019878A1 (en) Image processing device, image processing method, and image processing program
CN109472271B (zh) 印刷电路板图像轮廓提取方法及装置
CN107093174B (zh) 一种pcb设计缺陷检测方法
WO2017181724A1 (zh) 电子元件漏件检测方法和系统
CN106251333B (zh) 元件反件检测方法和系统
CN105976354A (zh) 基于颜色和梯度的元件定位方法和系统
WO2018068418A1 (zh) 元件安装状态检测方法和系统
WO2017071406A1 (zh) 金针类元件的引脚检测方法和系统
CN105205803A (zh) 显示面板缺陷检测方法
JP2018096908A (ja) 検査装置及び検査方法
CN114359253A (zh) 基于卷积神经网络的影像脏污检测方法及系统
JP5660861B2 (ja) 基板上の異物検査方法および異物検査装置
CN106815830B (zh) 图像的缺陷检测方法
CN115719326B (zh) Pcb板缺陷检测方法及装置
KR20200126921A (ko) 스켈러톤 웨이퍼 검사 방법
WO2017181722A1 (zh) 元件漏检方法和系统
WO2018068414A1 (zh) 色环电阻的检测方法、装置和自动光学检测系统
TWI496111B (zh) 針腳彎曲檢測方法
JP2009236550A (ja) 欠陥検出方法
CN114677348A (zh) 一种基于视觉的ic芯片缺陷的检测方法、系统和存储介质
CN111768357A (zh) 一种图像检测的方法及装置
KR101129708B1 (ko) 프로브 유닛의 블럭 베이스 검사장치, 및 이를 이용한 프로브 유닛의 블럭 베이스 검사방법
JP2014135007A (ja) フラットパネルディスプレイの自動ムラ検出装置および自動ムラ検出方法

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: 16918724

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: 16918724

Country of ref document: EP

Kind code of ref document: A1