WO2018072334A1 - 一种电子元件定位方法及装置 - Google Patents

一种电子元件定位方法及装置 Download PDF

Info

Publication number
WO2018072334A1
WO2018072334A1 PCT/CN2016/113600 CN2016113600W WO2018072334A1 WO 2018072334 A1 WO2018072334 A1 WO 2018072334A1 CN 2016113600 W CN2016113600 W CN 2016113600W WO 2018072334 A1 WO2018072334 A1 WO 2018072334A1
Authority
WO
WIPO (PCT)
Prior art keywords
template
sub
matching
electronic component
target
Prior art date
Application number
PCT/CN2016/113600
Other languages
English (en)
French (fr)
Inventor
林建民
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 WO2018072334A1 publication Critical patent/WO2018072334A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures

Definitions

  • the present invention relates to the field of automatic optical detection, and in particular, to an electronic component positioning method and apparatus.
  • Automated optical inspection is an essential part of the industrial manufacturing process, using optical methods to obtain the surface state of the finished product, and image processing to detect foreign matter or surface flaws.
  • the fault, leakage and reverse detection of electronic components is a common application in the field of circuit board defect detection.
  • the machine automatically scans the circuit board to acquire images, extracts partial images of each electronic component, and judges electronic components through image processing technology. Whether there are errors, leaks, and anti-defects, and finally display or mark the components with suspected defects for easy viewing and overhaul.
  • the precise positioning of electronic components is mainly obtained by template matching of color images, that is, the template image of the electronic components obtained by the workers during plate making is to be searched.
  • the area performs search matching to obtain positioning information of the electronic components.
  • the traditional positioning method is easily affected by the change of local regions in the template image (such as silk screen on electronic components, easy deformation parts on electronic components, etc.), resulting in deviations in the positioning results.
  • an object of the present invention is to provide an electronic component positioning method and apparatus, which can ensure the accuracy of the positioning result when performing positioning of the electronic component.
  • the invention provides an electronic component positioning method, comprising:
  • the preset electronic components to be positioned corresponds to the matching coordinates of the respective sub-template maps, the preset electronic components to be positioned, and the at least two submodules.
  • the relative position of the board image and the preset size of the electronic component to be positioned generate a positioning result.
  • the sub-template map is a picture of a stable area of the electronic component to be positioned in the plate picture of the plate, or a picture of the stable area of the electronic component to be positioned in the plate picture of the plate making .
  • the acquiring according to the position information of the at least two sub-template images corresponding to the electronic component to be located, the target image including the electronic component to be positioned from the image to be positioned corresponding to the sub-template image , Specifically:
  • the at least two sub-template maps are matched with the target image, and a set of multi-template matching degrees is generated according to the matching value of each target template in the target image, including :
  • a set of multiple template matching degrees is generated.
  • the at least two sub-template maps are matched with the target image, and a set of multi-template matching degrees is generated according to the matching value of each target template in the target image, including :
  • a set of multiple template matching degrees is generated.
  • the matching coordinates according to the respective template templates, the preset relative positions of the electronic components to be positioned and the at least two sub-template images, and the preset electronic components to be positioned Size generate positioning results, specifically:
  • the invention also provides an electronic component positioning device, comprising:
  • a target image obtaining unit configured to acquire, according to the position information of the at least two sub-template images corresponding to the electronic component to be located, the electronic component that includes the electronic component to be positioned from the image to be positioned corresponding to the sub-template image Target picture
  • a multi-template matching unit configured to match the at least two sub-template images with the target image, and generate a set of multi-template matching degrees according to matching values of the target image in each of the sub-template images ;
  • a matching degree selecting unit configured to select a target matching degree from the set of multiple template matching degrees, and acquire matching coordinates of each sub-template map corresponding to the target matching degree
  • a positioning unit configured to: according to the matching coordinates of the respective sub-template maps, the relative positions of the preset electronic components to be positioned and the at least two sub-template images, and the preset size of the electronic component to be positioned , generate positioning results.
  • the target picture obtaining unit specifically includes:
  • An enclosing area generating module configured to generate a minimum rectangular area surrounding the at least two sub-template images based on pre-selected position information of at least two sub-template images corresponding to the electronic component to be located;
  • an expansion module configured to expand the predetermined width of the minimum rectangular area to obtain a clipping window
  • an intercepting module configured to intercept, by the intercepting window, a target image that includes the electronic component to be located from the to-be-positioned image.
  • the multiple template matching unit specifically includes:
  • a first matching module configured to move each sub-template map on the target image, and obtain a matching value of each of the sub-template map and an area of the currently covered target image
  • a first multi-template matching calculation module configured to generate a current multi-template matching degree according to the matching value of each of the sub-template images; wherein, when the at least two sub-template templates are traversed by the target image After all regions, a set of multiple template matching degrees is generated.
  • the multiple template matching unit specifically includes:
  • a second matching module configured to move each sub-template map on the target image, and obtain a matching value of each of the sub-template map and an area of the currently covered target image
  • a geometric constraint calculation module configured to calculate a geometric constraint value between each sub-template image under the currently covered region according to relative position information between the respective sub-template images
  • a second multi-template matching calculation module configured to generate a current multi-template matching degree according to the matching value of each of the sub-template maps and the geometric constraint value; wherein, when the at least two sub-template templates are After traversing all areas of the target picture, a set of multiple template matching degrees is generated.
  • the positioning unit specifically includes:
  • a first central coordinate calculation module configured to calculate a center point coordinate of a region covered by each sub-template map according to matching coordinates of the respective sub-template maps
  • a second central coordinate calculation module configured to obtain, according to a central point coordinate of the area covered by each of the sub-template templates, a preset relative position of the electronic component to be positioned and the at least two sub-template images Describe the coordinates of the center point of the positioning electronic component;
  • a positioning module configured to generate a positioning result according to a center point coordinate of the electronic component to be positioned and a preset width and height of the electronic component to be positioned.
  • the electronic component positioning method and device provided by the present invention locates the electronic component to be positioned by using at least two sub-template images associated with the electronic component to be positioned and the target image at the same time, without directly adopting the entire electronic component to be positioned.
  • the template image is positioned, so that the influence of the easily changeable area on the electronic component to be positioned on the positioning result can be avoided, and the accuracy of the positioning result can be ensured.
  • FIG. 1 is a schematic flow chart of a method for positioning an electronic component according to an embodiment of the present invention.
  • FIG. 2 is a schematic diagram of a card picture of a plate making according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of an electronic component positioning device according to an embodiment of the present invention.
  • an embodiment of the present invention provides an electronic component positioning method for accurately positioning a position of an electronic component to be positioned from a target image including an electronic component to be positioned.
  • the electronic component positioning method can be performed by the electronic component positioning device, and includes the following steps:
  • the electronic component to be positioned in the image to be located before positioning the electronic component to be positioned in the image to be located, at least two sub-template diagrams related to the electronic component to be located are obtained.
  • the sub-template map is obtained from the card image of the plate making, and the to-be-positioned image is a picture of the same type of card as the plate-making board.
  • FIG. 2 is a schematic diagram of a card picture of a plate making according to an embodiment of the present invention.
  • 01 is a first sub-template diagram obtained from the board picture
  • 02 is a second sub-template diagram obtained from the board picture.
  • the sub-template diagram is a picture of a specific area of the electronic component to be located, or a picture on the board picture that is close to the area of the electronic component to be positioned.
  • the at least two sub-template diagrams are pictures of a stable area corresponding to the electronic component to be positioned in the plate-making picture of the plate, that is, the selection principle of the sub-template diagram is to avoid the easy Changed areas, such as silk screen changes on horizontal capacitors, plastic paper on transformers, etc., and select those areas that are more stable, that is, select areas that are not prone to change, because the easily changed areas may be on different boards. It will make a difference, which in turn will affect the accuracy of the final positioning.
  • the shape of the selected sub-template map is a rectangle.
  • the location information of each sub-template map needs to be acquired at the same time, such as the center point coordinate of each sub-template map. And coordinates of the vertices and the like, so that the target picture 03 including the electronic component to be positioned can be determined from the picture to be positioned corresponding to the sub-template image according to the location information.
  • a minimum rectangular area surrounding the at least two sub-template maps is generated based on pre-selected position information of at least two sub-template maps corresponding to the electronic components to be positioned.
  • the minimum rectangular area may be constructed by acquiring the coordinates of the vertices of each sub-template map and selecting the coordinates of the vertices located at the outermost side.
  • the minimum rectangular area is expanded by a predetermined width to generate a clipping window.
  • the target picture containing the electronic component to be located is cut out from the picture to be located by using the intercepting window.
  • the position information of the electronic component 04 to be positioned such as the center point coordinate (x, y) and the width w, height h, and the like.
  • the position information is selected by the user according to the actual shape of the electronic component to be positioned in the plate picture frame of the plate making, and the use thereof is to obtain the plate picture of the plate making, the sub-template image and the electronic component to be positioned. Relative position.
  • the coordinates of the center point of the first sub-template picture 01 on the board picture are (x 1_ori , y 1 — ori ), and the center of the second sheet template picture 02 on the board picture
  • the point coordinates are (x 2_ori , y 2_ori )
  • the relative positions of the respective sub-template maps and the electronic components to be positioned are:
  • S1021 Move each sub-template map on the target image, and obtain a matching value of each of the sub-template map and an area of the currently covered target image.
  • S1023 Generate a set of multiple template matching degrees after the at least two sub-template maps traverse all areas of the target picture.
  • a plurality of sub-template maps (assumed to be two) are simultaneously stacked on the upper left corner (or any corner) of the target image, and their matching values with the currently covered region are respectively calculated. Then keep the position of the second sub-template map unchanged, move the first sub-template map, and move the distance of one pixel to the right (or down) each time (automatically jump to the next line after the boundary, and change the movement Direction), wherein each time the first sub-template map is moved, the matching value of the first sub-template map and the currently covered region is calculated.
  • the second sub-template map is moved once, and the second sub-template map and the current overlay are recalculated.
  • the multi-template matching degree is calculated once every time the sub-template map is moved, so that after moving multiple times, a set of multi-template matching degrees is generated.
  • the matching value of the first sub-template and the currently covered region is R 1 (x 1 , y 1 )
  • the matching value of the second sub-template with the currently covered region is R 2 (x 2 , y 2 )
  • the method for calculating the matching value may be as follows:
  • T represents a sub-template map
  • I represents a region currently covered by the sub-template graph
  • x', y' are values of x and y coordinates in the sub-template graph T, respectively.
  • This type of method uses a multiplication operation between the template and the image, so a larger number indicates a higher degree of matching and a 0 indicates the worst matching effect.
  • This type of method matches the relative value of the template to its mean value and the correlation value of the image to its mean. 1 indicates a perfect match, -1 indicates a poor match, and 0 indicates no correlation (random sequence).
  • T '(x', y ' ) T (x', y ') - 1 / (w ⁇ h) ⁇ ⁇ x ", y" T (x ", y")
  • the target matching degree may be a maximum value or a minimum value of the set of multiple template matching degrees, and the specific case is determined according to the type of the matching algorithm selected. If the matching value of the selected matching algorithm is larger, the phase is represented. If the degree of similarity is greater, the maximum value of the set of multiple template matching degrees is selected as the target matching degree, and vice versa, the minimum value of the set of multiple template matching degrees is selected as the target matching degree.
  • each multi-template matching degree also corresponds to a set of matching coordinates, that is, a coordinate range of an area of the target picture covered by the at least two sub-templates.
  • the coordinate range of the area of the target picture covered by each sub-template is also acquired, that is, the matching coordinate (ie, (x 1 , y 1 ) corresponding to the target matching degree, (x) is acquired. 2 , y 2 ), where (x 1 , y 1 ) represents all coordinates of the response matrix obtained by the first sub-template diagram 01; (x 2 , y 2 ) represents the second sub-template diagram 02 Get all the coordinates of the response matrix).
  • S1041 Calculate, according to the matching coordinates of the respective sub-template maps, coordinate the center point coordinates of the area covered by each sub-template map.
  • the center point coordinates of the area currently covered by each sub-template map can be obtained, as can be obtained by the coordinates of two opposite vertices of the response matrix.
  • the coordinates of the center point of the area covered by the template template can be obtained.
  • a center point coordinate of a region corresponding to the first sheet template map is (x 1c , y 1c )
  • a center point coordinate of a region corresponding to the second sheet template map is (x 2c , y 2c )
  • the center point coordinates (x final , y final ) of the electronic component to be positioned are:
  • S1043 Generate a positioning result according to a center point coordinate of the electronic component to be positioned and a preset width and height of the electronic component to be positioned.
  • the position of the electronic component to be positioned may be determined according to the width w and the height h of the electronic component to be positioned.
  • the width w and height h of the locating electronic components are known at the time of plate making, and their sizes do not change.
  • the method for locating an electronic component locates the electronic component by using at least two sub-template images associated with the electronic component to be positioned and the target image at the same time, instead of directly adopting
  • the template image of the entire electronic component to be positioned is positioned, so that the change of the electronic component to be positioned can be avoided.
  • the influence of the area on the positioning result for example, the template image of the electronic component to be positioned acquired on the plate image of the plate is not exactly the same as the electronic component to be positioned on the actual card to be positioned, such as silk screen on the electronic component, electronic
  • the deformable parts on the components may be slightly different
  • step S102 may further include:
  • S1025 Calculate geometric constraint values between the sub-template images under the currently covered region according to the relative position information between the respective sub-template images.
  • S1027 Generate a set of multi-template matching degrees after the at least two sub-template maps traverse all areas of the target picture.
  • the multi-template matching degree when calculating the multi-template matching degree, the geometric distance constraint relationship between the multiple template images is also considered. Specifically, the multi-template matching degree is calculated as follows:
  • R 1 (x, y) and R 2 (x, y) respectively correspond to the matching values of the first sub-template template 01 and the second sub-template 02
  • ⁇ x and ⁇ y respectively represent the first set set at the time of plate making.
  • is a coefficient that represents the weight of the geometric constraint. The smaller the value, the smaller the influence of the geometric constraint.
  • this will reduce the matching degree of the more template matching degree R(x 1 , y 1 , x 2 , y 2 ), that is, the effect of the offset of the penalty sub-template positioning.
  • the positive/negative of ⁇ is related to the matching method of template matching.
  • ⁇ 0 squared matching/standard squared matching is used, ⁇ 0; when using correlation matching/standard correlation matching, ⁇ >0.
  • the geometric sub-template positioning error is punctured by introducing a geometric distance constraint, so that the positioning result is more accurate.
  • an embodiment of the present invention further provides an electronic component positioning apparatus 100, including:
  • the target image obtaining unit 10 is configured to acquire the electronic component to be located from the to-be-positioned image corresponding to the sub-template image based on the pre-selected position information of the at least two sub-template images corresponding to the electronic component to be located.
  • the target picture is configured to acquire the electronic component to be located from the to-be-positioned image corresponding to the sub-template image based on the pre-selected position information of the at least two sub-template images corresponding to the electronic component to be located.
  • the multi-template matching unit 20 is configured to match the at least two sub-template images with the target image, and generate a set of multi-template matching according to the matching value of each target template in the target image. degree.
  • the matching degree selecting unit 30 is configured to select a target matching degree from the set of multiple template matching degrees, and acquire matching coordinates of each sub-template map corresponding to the target matching degree;
  • the locating unit 40 is configured to: according to the matching coordinates of the respective sub-template maps, the relative positions of the preset electronic components to be positioned and the at least two sub-template images, and the preset electronic components to be positioned Size, generate positioning results.
  • the target picture obtaining unit 10 specifically includes:
  • the enclosing area generating module 11 is configured to generate a minimum rectangular area surrounding the at least two sub-template images based on the pre-selected position information of the at least two sub-template images corresponding to the electronic component to be located.
  • the expansion module 12 is configured to expand the predetermined width of the minimum rectangular area to obtain a clipping window.
  • the intercepting module 13 is configured to extract, from the to-be-positioned picture, a target picture that includes the electronic component to be located by using the intercepting window.
  • the multiple template matching unit 20 specifically includes:
  • the first matching module 21 is configured to move each sub-template map on the target image, and obtain a matching value of each of the sub-template map and an area of the currently covered target image.
  • the first multi-template matching calculation module 22 is configured to generate a current multi-template matching degree according to the matching value of each of the sub-template images; wherein, when the at least two sub-template templates all traverse the target image After all regions have been generated, a set of multiple template matching degrees is generated.
  • the multiple template matching unit 20 specifically includes:
  • the second matching module 23 is configured to move each sub-template map on the target image, and obtain a matching value of each of the sub-template map and an area of the currently covered target image.
  • the geometric constraint calculation module 24 is configured to calculate the current coverage according to the relative position information between the respective sub-template images Under the area, the geometric constraint values between each sub-template map.
  • the second multi-template matching calculation module 25 is configured to generate a current multi-template matching degree according to the matching value of each of the sub-template maps and the geometric constraint value; wherein, at least two sub-template templates are After traversing all the regions of the target picture, a set of multiple template matching degrees is generated.
  • the positioning unit 40 specifically includes:
  • the first central coordinate calculation module 41 is configured to calculate, according to the matching coordinates of the respective sub-template maps, the coordinates of the center point of the region covered by each sub-template map.
  • the second central coordinate calculation module 42 is configured to obtain, according to a central point coordinate of the area covered by each of the sub-template templates, a preset relative position of the electronic component to be positioned and the at least two sub-template images The center point coordinate of the electronic component to be positioned.
  • the positioning module 43 is configured to generate a positioning result according to the center point coordinates of the electronic component to be positioned and the preset width and height of the electronic component to be positioned.
  • the electronic component positioning device 100 provided by the present invention can locate the electronic component by using the method of matching at least two sub-template images and the target image at the same time, instead of directly using the template image of the entire electronic component to be positioned, thereby avoiding The influence of the changeable area on the electronic component to be positioned on the positioning result ensures the accuracy of the positioning result.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Image Analysis (AREA)

Abstract

一种电子元件定位方法,包括:基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片(S101);将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度(S102);从所述的一组多模板匹配度中选取目标匹配度,并获取对应于所述目标匹配度的各个子模板图的匹配坐标(S103);根据所述匹配坐标、待定位电子元件与子模板图的相对位置及所述待定位电子元件的尺寸,生成定位结果(S104)。一种电子元件定位装置,保证定位结果的准确性。

Description

一种电子元件定位方法及装置 技术领域
本发明涉及自动光学检测领域,尤其涉及一种电子元件定位方法及装置。
背景技术
自动光学检测是工业制作过程的必要环节,利用光学方式取得成品的表面状态,以影像处理来检测异物或表面瑕疵。电子元件的错、漏、反检测是电路板缺陷检测领域中的一种常见应用,机器通过摄像头自动扫描电路板获取图像,提取每个电子元件的局部图像,并通过图像处理技术,判断电子元件是否存在错、漏、反缺陷,最后将疑似缺陷的元件显示或标记出来,方便查看与检修。
在传统的自动光学检测(Automatic Optic Inspection,AOI)系统中,电子元件的精确定位主要是通过彩色图像的模板匹配得到的,也即是通过工人制版时得到的电子元件的模板图片在待搜索的区域进行搜索匹配,以得到电子元件的定位信息。但是传统的定位方法容易受到模板图像中局部区域(如电子元件上的丝印,电子元件上易形变部位等)变化的影响,使得定位的结果出现偏差。
发明内容
针对上述问题,本发明的目的在于提供一种电子元件定位方法及装置,在进行电子元件的定位时,可保证定位结果的准确性。
本发明提供了一种电子元件定位方法,包括:
基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片;
将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度;
从所述的一组多模板匹配度中选取目标匹配度,并获取对应于所述目标匹配度的各个子模板图的匹配坐标;
根据所述各个子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模 板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果。
优选地,所述子模板图为位于制版的板卡图片中的待定位电子元件的稳定区域的图片,或者为所述制版的板卡图片中的靠近所述待定位电子元件的稳定区域的图片。
优选地,所述基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片,具体为:
基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,生成包围所述的至少两张子模板图的最小矩形区域;
对所述最小矩形区域扩展预定的宽度,生成截取窗口;
以所述截取窗口从待定位图片上截取出包含所述待定位电子元件的目标图片。
优选地,所述将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度,包括:
在目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
根据每个所述子模板图的匹配值,生成当前的多模板匹配度;
在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
优选地,所述将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度,包括:
在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
根据各个子模板图之间的相对位置信息,计算在当前覆盖的区域下,各个子模板图之间的几何约束值;
根据每个所述子模板图的匹配值及所述几何约束值,生成当前的多模板匹配度;
在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
优选地,所述根据所述各张子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果,具体为:
根据所述各个子模板图的匹配坐标,计算得到各张子模板图覆盖的区域的中心点坐标;
根据所述各张子模板图覆盖的区域的中心点坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,获得所述待定位电子元件的中心点坐标;
根据所述待定位电子元件的中心点坐标及预设的所述待定位电子元件的宽和高,生成定 位结果。
本发明还提供了一种电子元件定位装置,包括:
目标图片获取单元,用于基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片;
多模板匹配单元,用于将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度;
匹配度选取单元,用于从所述的一组多模板匹配度中选取目标匹配度,并获取对应于所述目标匹配度的各个子模板图的匹配坐标;
定位单元,用于根据所述各个子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果。
优选地,所述目标图片获取单元具体包括:
包围区域生成模块,用于基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,生成包围所述的至少两张子模板图的最小矩形区域;
扩展模块,用于对所述最小矩形区域扩展预定的宽度,获取截取窗口;
截取模块,用于以所述截取窗口从待定位图片上截取出包含所述待定位电子元件的目标图片。
优选地,所述多模板匹配单元具体包括:
第一匹配模块,用于在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
第一多模板匹配计算模块,用于根据每个所述子模板图的匹配值,生成当前的多模板匹配度;其中,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
优选地,所述多模板匹配单元具体包括:
第二匹配模块,用于在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
几何约束计算模块,用于根据各个子模板图之间的相对位置信息,计算在当前覆盖的区域下,各个子模板图之间的几何约束值;
第二多模板匹配计算模块,用于根据每个所述子模板图的匹配值及所述几何约束值,生成当前的多模板匹配度;其中,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
优选地,所述定位单元具体包括:
第一中心坐标计算模块,用于根据所述各个子模板图的匹配坐标,计算得到各张子模板图覆盖的区域的中心点坐标;
第二中心坐标计算模块,用于根据所述各张子模板图覆盖的区域的中心点坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,获得所述待定位电子元件的中心点坐标;
定位模块,用于根据所述待定位电子元件的中心点坐标及预设的所述待定位电子元件的宽和高,生成定位结果。
本发明提供的电子元件定位方法及装置,通过利用与待定位电子元件相关的至少两个子模板图与目标图片同时进行匹配的方法对待定位电子元件进行定位,而不直接采用整个待定位电子元件的模板图像进行定位,因此可避免所述待定位电子元件上的易变化区域对定位结果的影响,保证定位结果的准确性。
附图说明
为了更清楚地说明本发明的技术方案,下面将对实施方式中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施方式,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的电子元件定位方法的流程示意图。
图2是本发明实施例提供的一种制版的板卡图片的示意图。
图3是本发明实施例提供的电子元件定位装置的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
请参阅图1,本发明实施例提供一种电子元件定位方法,用于从包含有待定位电子元件的目标图片中精确定位出待定位电子元件的位置。其中,所述电子元件定位方法可由电子元件定位装置来执行,并包括如下步骤:
S101,基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应 于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片。
在本发明实施例中,在对位于待定位图片内的待定位电子元件进行定位之前,需先获取与所述待定位电子元件相关的至少两张子模板图。其中,所述子模板图从制版的板卡图片上截取得到,所述待定位图片则是与制版的板卡同一类型的板卡的图片。
如图2所示,图2为本发明实施例提供的一种制版的板卡图片的示意图。
其中,01为从所述板卡图片上获取的第一张子模板图,02为从所述板卡图片上获得的第二张子模板图。在本发明实施例中,所述子模板图为待定位电子元件的特定区域的图片,或者是所述板卡图片上的靠近所述待定位电子元件的区域的图片。特别的,所述的至少两张子模板图为位于制版的板卡图片中的与所述待定位电子元件对应的稳定区域的图片,即所述子模板图的选取原则是要避开那些容易变化的区域,如卧式电容上变化的丝印,变压器上变化的塑料纸等,而选择那些较为稳定的区域,即选择那些不容易出现变化的区域,因为容易变化的区域在不同板卡上可能会有所区别,进而会影响最终定位的精度。
在本发明实施例中,一般地,选取的所述子模板图的形状为矩形,在选取的时候,还需要同时获取每张子模板图的位置信息,如每张子模板图的中心点坐标及各个顶点的坐标等,从而可以根据所述位置信息从相应于所述子模板图的待定位图片中确定包含所述待定位电子元件的目标图片03。
具体地,可通过如下方法进行确定:
首先,基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,生成包围所述的至少两张子模板图的最小矩形区域。
其中,所述最小矩形区域可通过获取各个子模板图的顶点的坐标,并选取位于最外侧的顶点的坐标来构造得到。
然后,对所述最小矩形区域扩展预定的宽度,生成截取窗口。
例如,沿所述最小矩形区域的四条边向外扩展30或N个像素,生成截取窗口03。
最后,以所述截取窗口从待定位图片上截取出包含所述待定位电子元件的目标图片。
需要说明的是,在制版的板卡图片上,还需确定待定位电子元件04的位置信息,如中心点坐标(x,y)以及宽w、高h等。这些位置信息由用户根据所述待定位电子元件的实际形状在所述制版的板卡图片框选得到,其用处是获取制版的板卡图片上,所述子模板图与所述待定位电子元件的相对位置。
例如,假设所述第一张子模板图片01在所述板卡图片上的中心点坐标是(x1_ori,y1_ori),所述第二张子模板图片02在所述板卡图片上的中心点坐标是(x2_ori,y2_ori),则各个子模板图与所述待定位电子元件的相对位置为:
Figure PCTCN2016113600-appb-000001
其中,xcenter=(x1_ori+x2_ori)/2,ycenter=(y1_ori+y2_ori)/2。
S102,将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度。
具体地,可包括如下步骤:
S1021,在目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值。
S1022,根据每个所述子模板图的匹配值,生成当前的多模板匹配度。
S1023,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
具体地:
首先,同时将多张子模板图(假设为两张)叠放于所述目标图片的左上角(或任意一个角),并分别计算其与当前覆盖的区域的匹配值。然后保持第二张子模板图位置不变,移动第一张所述子模板图,且每次向右(或向下)移动一个像素的距离(到边界后自动跳转下一行,并改变移动方向),其中,每移动一次第一张子模板图,都计算一次第一张子模板图与当前覆盖的区域的匹配值。
接着,在所述第一张子模板图遍历所述目标图片并回到左上角的位置后,移动一次所述第二张子模板图,并重新计算所述第二张子模板图与当前覆盖的区域的匹配值,此后,再让所述第一张子模板图再遍历一次目标图片,依此循环,直到所述第二张子模板图移动到所述目标图片的右下角。
在本发明实施例中,每移动一次子模板图都会计算一次多模板匹配度,这样,在移动多次后,将生成一组多模板匹配度。其中,假设第一张子模板与当前覆盖的区域的匹配值为R1(x1,y1),第二张子模板与当前覆盖的区域的匹配值为R2(x2,y2),则当前的多模板匹配度为R(x1,y1,x2,y2)=R1(x1,y1)+R2(x2,y2)。
在本发明实施例中,计算所述匹配值可有如下几种方法:
(1)平方差匹配
Figure PCTCN2016113600-appb-000002
其中,R(x,y)为匹配值,T表示子模板图,I表示所述子模板图当前覆盖的区域,x′,y′分别是子模板图T中x、y坐标的值。
(2)标准平方差匹配
Figure PCTCN2016113600-appb-000003
(3)相关匹配
这类方法采用模板和图像间的乘法操作,所以较大的数表示匹配程度较高,0标识最坏的匹配效果。
Figure PCTCN2016113600-appb-000004
(4)标准相关匹配
Figure PCTCN2016113600-appb-000005
(5)相关匹配method=CV_TM_CCOEFF
这类方法将模版对其均值的相对值与图像对其均值的相关值进行匹配,1表示完美匹配,-1表示糟糕的匹配,0表示没有任何相关性(随机序列)。
Figure PCTCN2016113600-appb-000006
其中:
T′(x′,y′)=T(x′,y′)-1/(w·h)·∑x″,y″T(x″,y″)
I′(x+x′,y+y′)=I(x+x′,y+y′)-1/(w·h)·∑x″,y″I(x+x″,y+y″)
(6)标准相关匹配:
Figure PCTCN2016113600-appb-000007
当然,还可有其他的匹配算法,本发明不做具体限定。
S103,从所述的一组多模板匹配度中选取目标匹配度,并获取与所述目标匹配度对应的匹配位置信息。
在本发明实施例中,所述的目标匹配度可以是所述的一组多模板匹配度中的最大值或最小值,其具体情况根据选取的匹配算法的类型决定。若选取的匹配算法的匹配值越大代表相 似度越大,则选取所述的一组多模板匹配度中的最大值作为目标匹配度,反之,则选取所述的一组多模板匹配度中的最小值作为目标匹配度。
在本发明实施例中,每个多模板匹配度还对应于一组匹配坐标,即所述的至少两个子模板覆盖的目标图片的区域的坐标范围。在获得所述目标匹配度后,还需同时获取每个子模板覆盖的目标图片的区域的坐标范围,即获取匹配坐标(即与所述目标匹配度对应的(x1,y1),(x2,y2),其中,这里的(x1,y1)表示第一张子模板图01所得到的响应矩阵的所有坐标;(x2,y2)表示第二张子模板图02所得到的响应矩阵的所有坐标)。
S104,根据所述匹配位置信息及预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,生成定位结果。
具体地:
S1041,根据所述各个子模板图的匹配坐标,计算得到各张子模板图覆盖的区域的中心点坐标。
在本发明实施例中,在获得所述匹配坐标后,即可获得各张子模板图当前覆盖的区域的中心点坐标,如可通过所述响应矩阵的两个相对的顶点的坐标求得各张子模板图覆盖的区域的中心点坐标。
S1042,根据所述中心点坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,获得所述待定位电子元件的中心点坐标
在本发明实施例中,假设与所述第一张子模板图对应的区域的中心点坐标为(x1c,y1c),与所述第二张子模板图对应的区域的中心点坐标为(x2c,y2c),则所述待定位电子元件的中心点坐标(xfinal,yfinal)为:
Figure PCTCN2016113600-appb-000008
S1043,根据所述待定位电子元件的中心点坐标及预设的所述待定位电子元件的宽和高,生成定位结果
在本发明实施例中,在获得所述待定位电子元件的中心点坐标后,即可以根据所述待定位电子元件的宽w和高h确定出所述待定位电子元件的位置,其中,所述待定位电子元件的宽w和高h在制版时已知,它们的大小不会发生变化。
综上所述,本发明实施例提供的电子元件的定位方法,通过利用与待定位电子元件相关的至少两个子模板图与目标图片同时进行匹配的方法对待定位电子元件进行定位,而不直接采用整个待定位电子元件的模板图像进行定位,因此可避免所述待定位电子元件上的易变化 区域对定位结果的影响(例如,在制版的板卡图像上获取的待定位电子元件的模板图与实际的需要定位的板卡上待定位电子元件不完全相同,例如电子元件上的丝印,电子元件上易形变部位可能会略有不同),保证定位结果的准确性。
为了便于对本发明的理解,下面将对本发明的一些优选实施例做更进一步的描述。
优选地,步骤S102还可包括:
S1024,根据预设的移动规则在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值。
S1025,根据各个子模板图之间的相对位置信息,计算在当前覆盖的区域下,各个子模板图之间的几何约束值。
S1026,根据每个所述子模板图的匹配值及所述几何约束值,生成当前的多模板匹配度。
S1027,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
在本优选实施例中,在计算所述多模板匹配度的时候,还同时考虑多个模板图像之间的几何距离约束关系,具体地,所述多模板匹配度的计算公式如下:
Figure PCTCN2016113600-appb-000009
其中R1(x,y)和R2(x,y)分别对应第一张子模板图01、第二张子模板图02的匹配值,Δx,Δy分别表示制版时设定的第一张子模板图01和第二张子模板图02中预定位置(如左上角坐标)的差值。λ是一个系数,该系数表示几何约束的权重,其值越小,表示几何约束影响越小。
在本优选实施例中,通过引入几何约束值
Figure PCTCN2016113600-appb-000010
来惩罚个别子模板图定位出错的情况,其原理是:在原始状态下面,各个子模板之间的位置关系是确定的,也即是说当定位精准的话,应该有下面的关系:
Figure PCTCN2016113600-appb-000011
如果有某个子模板图定位出现偏移,那么就会使得上式不再成立,这个时候就会使得:
Figure PCTCN2016113600-appb-000012
也就是说,这会使得多模板匹配度R(x1,y1,x2,y2)的匹配度下降,也即达到惩罚子模板定位出现偏移的作用。
其中λ的正负与模板匹配的匹配度评价方法相关,当使用平方差匹配/标准平方差匹配时,λ<0;当使用相关匹配/标准相关匹配时,λ>0。
本优选实施例中,通过引入几何距离约束来惩罚个别子模板定位出错的情况,使得定位结果更准确。
请参阅图3,本发明实施例还提供一种电子元件定位装置100,包括:
目标图片获取单元10,用于基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片。
多模板匹配单元20,用于将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度。
匹配度选取单元30,用于从所述的一组多模板匹配度中选取目标匹配度,并获取对应于所述目标匹配度的各个子模板图的匹配坐标;
定位单元40,用于根据所述各个子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果。
优选地,所述目标图片获取单元10具体包括:
包围区域生成模块11,用于基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,生成包围所述的至少两张子模板图的最小矩形区域。
扩展模块12,用于对所述最小矩形区域扩展预定的宽度,获取截取窗口。
截取模块13,用于以所述截取窗口从待定位图片上截取出包含所述待定位电子元件的目标图片。
优选地,所述多模板匹配单元20具体包括:
第一匹配模块21,用于在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值。
第一多模板匹配计算模块22,用于根据每个所述子模板图的匹配值,生成当前的多模板匹配度;其中,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
优选地,所述多模板匹配单元20具体包括:
第二匹配模块23,用于在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值。
几何约束计算模块24,用于根据各个子模板图之间的相对位置信息,计算在当前覆盖的 区域下,各个子模板图之间的几何约束值。
第二多模板匹配计算模块25,用于根据每个所述子模板图的匹配值及所述几何约束值,生成当前的多模板匹配度;其中,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
优选地,所述定位单元40具体包括:
第一中心坐标计算模块41,用于根据所述各个子模板图的匹配坐标,计算得到各张子模板图覆盖的区域的中心点坐标。
第二中心坐标计算模块42,用于根据所述各张子模板图覆盖的区域的中心点坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,获得所述待定位电子元件的中心点坐标。
定位模块43,用于根据所述待定位电子元件的中心点坐标及预设的所述待定位电子元件的宽和高,生成定位结果。
本发明提供的电子元件定位装置100,通过利用至少两个子模板图与目标图片同时进行匹配的方法对待定位电子元件进行定位,而不直接采用整个待定位电子元件的模板图像进行定位,因此可避免所述待定位电子元件上的易变化区域对定位结果的影响,保证定位结果的准确性。
以上所揭露的仅为本发明一种较佳实施例而已,当然不能以此来限定本发明之权利范围,本领域普通技术人员可以理解实现上述实施例的全部或部分流程,并依本发明权利要求所作的等同变化,仍属于发明所涵盖的范围。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于一计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)或随机存储记忆体(Random Access Memory,RAM)等。

Claims (10)

  1. 一种电子元件定位方法,其特征在于,包括:
    基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片;
    将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度;
    从所述的一组多模板匹配度中选取目标匹配度,并获取对应于所述目标匹配度的各个子模板图的匹配坐标;
    根据所述各个子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果。
  2. 根据权利要求1所述的电子元件定位方法,其特征在于,所述子模板图为位于制版的板卡图片中的待定位电子元件的稳定区域的图片,或者为所述制版的板卡图片中的靠近所述待定位电子元件的稳定区域的图片。
  3. 根据权利要求1所述的电子元件定位方法,其特征在于,所述基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片,具体为:
    基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,生成包围所述的至少两张子模板图的最小矩形区域;
    对所述最小矩形区域扩展预定的宽度,生成截取窗口;
    以所述截取窗口从待定位图片上截取出包含所述待定位电子元件的目标图片。
  4. 根据权利要求1所述的电子元件定位方法,其特征在于,所述将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度,包括:
    在目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
    根据每个所述子模板图的匹配值,生成当前的多模板匹配度;
    在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
  5. 根据权利要求1所述的电子元件定位方法,其特征在于,所述将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度,包括:
    在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
    根据各个子模板图之间的相对位置信息,计算在当前覆盖的区域下,各个子模板图之间的几何约束值;
    根据每个所述子模板图的匹配值及所述几何约束值,生成当前的多模板匹配度;
    在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
  6. 根据权利要求1所述的电子元件定位方法,其特征在于,所述根据所述各张子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果,具体为:
    根据所述各个子模板图的匹配坐标,计算得到各张子模板图覆盖的区域的中心点坐标;
    根据所述各张子模板图覆盖的区域的中心点坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,获得所述待定位电子元件的中心点坐标;
    根据所述待定位电子元件的中心点坐标及预设的所述待定位电子元件的宽和高,生成定位结果。
  7. 一种电子元件定位装置,其特征在于,包括:
    目标图片获取单元,用于基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,从相应于所述子模板图的待定位图片中获取包含所述待定位电子元件的目标图片;
    多模板匹配单元,用于将所述的至少两张子模板图与所述目标图片进行匹配,并根据每张所述子模板图在所述目标图片的匹配值,生成一组多模板匹配度;
    匹配度选取单元,用于从所述的一组多模板匹配度中选取目标匹配度,并获取对应于所述目标匹配度的各个子模板图的匹配坐标;
    定位单元,用于根据所述各个子模板图的匹配坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置及预设的所述待定位电子元件的尺寸,生成定位结果。
  8. 根据权利要求7所述的电子元件定位装置,其特征在于,所述目标图片获取单元具体包括:
    包围区域生成模块,用于基于预先选取的与待定位电子元件对应的至少两张子模板图的位置信息,生成包围所述的至少两张子模板图的最小矩形区域;
    扩展模块,用于对所述最小矩形区域扩展预定的宽度,获取截取窗口;
    截取模块,用于以所述截取窗口从待定位图片上截取出包含所述待定位电子元件的目标图片。
  9. 根据权利要求7所述的电子元件定位装置,其特征在于,所述多模板匹配单元具体包括:
    第二匹配模块,用于在所述目标图片上移动各个子模板图,并获取每个所述子模板图与当前覆盖的目标图片的区域的匹配值;
    几何约束计算模块,用于根据各个子模板图之间的相对位置信息,计算在当前覆盖的区域下,各个子模板图之间的几何约束值;
    第二多模板匹配计算模块,用于根据每个所述子模板图的匹配值及所述几何约束值,生成当前的多模板匹配度;其中,在当所述的至少两张子模板图均遍历所述目标图片的所有区域后,生成一组多模板匹配度。
  10. 根据权利要求7所述的电子元件定位装置,其特征在于,所述定位单元具体包括:
    第一中心坐标计算模块,用于根据所述各个子模板图的匹配坐标,计算得到各张子模板图覆盖的区域的中心点坐标;
    第二中心坐标计算模块,用于根据所述各张子模板图覆盖的区域的中心点坐标、预设的所述待定位电子元件与所述的至少两张子模板图的相对位置,获得所述待定位电子元件的中心点坐标;
    定位模块,用于根据所述待定位电子元件的中心点坐标及预设的所述待定位电子元件的宽和高,生成定位结果。
PCT/CN2016/113600 2016-10-20 2016-12-30 一种电子元件定位方法及装置 WO2018072334A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610918054.5A CN106503737B (zh) 2016-10-20 2016-10-20 一种电子元件定位方法及装置
CN201610918054.5 2016-10-20

Publications (1)

Publication Number Publication Date
WO2018072334A1 true WO2018072334A1 (zh) 2018-04-26

Family

ID=58319352

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/113600 WO2018072334A1 (zh) 2016-10-20 2016-12-30 一种电子元件定位方法及装置

Country Status (2)

Country Link
CN (1) CN106503737B (zh)
WO (1) WO2018072334A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107330879A (zh) * 2017-06-21 2017-11-07 广州视源电子科技股份有限公司 一种器件的检查方法、装置、设备及存储介质
CN109509165B (zh) * 2017-09-11 2021-01-29 凌云光技术股份有限公司 图像定位区域选取方法及装置
CN108709500B (zh) * 2018-05-11 2020-08-11 佛山科学技术学院 一种电路板元件定位匹配方法
CN110751682B (zh) * 2019-10-28 2022-07-05 普联技术有限公司 一种提取和标识图像的方法、装置、终端设备及存储介质
CN111889222A (zh) * 2020-08-03 2020-11-06 湖南大奇智能科技有限公司 基于视觉伺服的摇床导流控制系统和摇床导流控制方法
CN112634227A (zh) * 2020-12-21 2021-04-09 广州镭晨智能科技有限公司 Pcb拼板的检测标识方法、装置、电子设备以及存储介质
CN116385353B (zh) * 2023-02-10 2024-01-30 南通大学 一种摄像头模组异常检测方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103337068A (zh) * 2013-06-04 2013-10-02 华中科技大学 空间关系约束的多子区匹配方法
CN104463178A (zh) * 2014-12-29 2015-03-25 广州视源电子科技股份有限公司 电子元件识别方法和系统
CN105069799A (zh) * 2015-08-13 2015-11-18 深圳市华汉伟业科技有限公司 一种角点定位方法及装置
CN105260740A (zh) * 2015-09-23 2016-01-20 广州视源电子科技股份有限公司 一种元件识别方法及装置
CN105354816A (zh) * 2015-09-24 2016-02-24 广州视源电子科技股份有限公司 一种电子元件定位方法及装置
CN105404900A (zh) * 2015-12-22 2016-03-16 广州视源电子科技股份有限公司 一种并排二极管的定位方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103337068A (zh) * 2013-06-04 2013-10-02 华中科技大学 空间关系约束的多子区匹配方法
CN104463178A (zh) * 2014-12-29 2015-03-25 广州视源电子科技股份有限公司 电子元件识别方法和系统
CN105069799A (zh) * 2015-08-13 2015-11-18 深圳市华汉伟业科技有限公司 一种角点定位方法及装置
CN105260740A (zh) * 2015-09-23 2016-01-20 广州视源电子科技股份有限公司 一种元件识别方法及装置
CN105354816A (zh) * 2015-09-24 2016-02-24 广州视源电子科技股份有限公司 一种电子元件定位方法及装置
CN105404900A (zh) * 2015-12-22 2016-03-16 广州视源电子科技股份有限公司 一种并排二极管的定位方法及装置

Also Published As

Publication number Publication date
CN106503737B (zh) 2019-03-05
CN106503737A (zh) 2017-03-15

Similar Documents

Publication Publication Date Title
WO2018072334A1 (zh) 一种电子元件定位方法及装置
WO2017092427A1 (zh) 一种电子元件定位方法及装置
TWI485650B (zh) 用於多相機校準之方法及配置
TWI604197B (zh) 用於測量多層結構的層之間疊對的技術
CN113688807B (zh) 一种自适应缺陷检测方法、装置、识别系统及存储介质
JP5699788B2 (ja) スクリーン領域検知方法及びシステム
US7194709B2 (en) Automatic alignment of integrated circuit and design layout of integrated circuit to more accurately assess the impact of anomalies
WO2017181724A1 (zh) 电子元件漏件检测方法和系统
WO2017215241A1 (zh) 电路板元件漏件检测方法和系统
KR102649038B1 (ko) 비전 시스템으로 이미지에서 라인을 찾기 위한 시스템 및 방법
WO2017177717A1 (zh) 基于颜色和梯度的元件定位方法和系统
CA2507174A1 (en) Method of registering and aligning multiple images
US9851714B2 (en) Method of inspecting a specimen and system thereof
TW201915792A (zh) 智慧型的缺陷校正系統與其實施方法
WO2018006566A1 (zh) 视图调整方法和系统
CN104423142B (zh) 用于光学邻近校正模型的校准数据收集方法和系统
WO2017050088A1 (zh) 一种电子元件定位方法及装置
CN114998097A (zh) 图像对齐方法、装置、计算机设备和存储介质
Zhang et al. A new algorithm for accurate and automatic chessboard corner detection
JP2006214816A (ja) 半導体検査装置
KR20160014548A (ko) Sem 이미지들과 cad 데이터의 레지스트레이션
JP2013140468A (ja) パターンマッチング装置、検査システム、及びコンピュータプログラム
WO2024012463A1 (zh) 一种定位方法及装置
JP2004536300A5 (zh)
CN111354038A (zh) 锚定物检测方法及装置、电子设备及存储介质

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: 16919397

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 07/10/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16919397

Country of ref document: EP

Kind code of ref document: A1