CN102930266B - Method for locating Mark points on PCB (printed circuit board) by utilizing outline gravity center method - Google Patents
Method for locating Mark points on PCB (printed circuit board) by utilizing outline gravity center method Download PDFInfo
- Publication number
- CN102930266B CN102930266B CN201210369187.3A CN201210369187A CN102930266B CN 102930266 B CN102930266 B CN 102930266B CN 201210369187 A CN201210369187 A CN 201210369187A CN 102930266 B CN102930266 B CN 102930266B
- Authority
- CN
- China
- Prior art keywords
- mark
- gravity
- profile
- point
- mark point
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a method for locating Mark points on a PCB (printed circuit board) by utilizing an outline gravity center method. The method is characterized by comprising the following steps: on the basis of taking a Mark point extracted in initial programming as a datum point, taking the outline gravity center of the Mark point of the board to be detected, comparing the outline gravity center of the Mark point with the gravity center of the datum point, and performing PCB compensation by virtue of offset between the two gravity centers. The method is used for performing the PCB compensation by extracting the outline gravity center of the Mark point on the PCB and utilizing the offset between the gravity centers of the datum Mark point and the Mark point to be detected. The gravity centers are obtained by binary images and are easy and feasible to obtain. The algorithm is independent from the shapes of the Mark points, is not influenced by the intensity of illumination, achieves sub-pixel-level precision, and is high in compensation precision.
Description
Technical field
The present invention relates to PCB welding field of visual inspection, particularly about a kind of method of ordering with profile gravity model appoach location pcb board Mark, make, in the time locating, proofread and correct pcb board, can effectively improve compensation precision.
Background technology
Along with the development of computer digital image treatment technology and Dynamic Non-Destruction Measurement, detect with recognition technology day by day towards intellectuality, high reliability, high-speed automated online detection future development now.Computer vision technique and reliability engineering are two large technical foundation of modern measure.Machine vision is successfully applied in industrial detection field, wherein comprise the automatic detection to PCB (printed circuit board), realize automatic location to circuit board and the correction of benchmark, aborning accurate arrangement components and parts are had to important effect, this is because in the time of the components and parts that weld on pcb board, owing to putting pcb board at every turn, the position of plate is easily offset, so just can not weld standard when solder pad.Can to weld standard when making to weld, must to take certain measure that pcb board is accurately located.
In the pcb board design phase, deviser can make mark point conventionally as positioning datum, has had position relationship accurately, just can carry out basis for follow-up effective detection.The Mark point of PCB is that in board design, PCB is applied to the location recognition point on automatic placement machine, is also gauge point or unique point.
The ultimate principle of vision-based detection pcb board is: after image is processed, judge the position of mark according to the statistics of pixel, then the identification point detecting and standard setting point are compared, if find error, compensated, improve the positioning precision of pcb board.
The existing image processing method that pcb board is detected is automatically images match method.What but this method can only processed pixels rank, can not process sub-pixel other.And the method is subject to the impact of ambient light photograph, in the time that external light source changes greatly, the method there will be the situation of wrong report.
Summary of the invention
For the problems referred to above, the object of this invention is to provide a kind of method of ordering with profile gravity model appoach location pcb board Mark.Use the method can improve the positional accuracy of pcb board, improve welding precision.
For achieving the above object, the present invention takes following technical scheme: a kind of method of ordering with profile gravity model appoach location pcb board Mark, feature is: taking the Mark point that extracts when the initial programming as reference point, by the Mark point contouring center of gravity of drafting board to be checked, compare with reference point center of gravity, utilize the skew between center of gravity to carry out pcb board compensation.
Get the Mark dot profile center of gravity of drafting board to be checked, and the concrete grammar that compares, compensates with benchmark Mark point is as follows:
1) take lower pcb board to be detected with camera and contain the image that Mark is ordered, be presented on computer display screen, the sectional drawing control then carrying with tin soldering robot intercepts and contains the area image that Mark is ordered in current visual field;
2) tin soldering robot carries out binary conversion treatment to the Mark point area image intercepting, and obtains binary image;
3) binary image is extracted to the real profile that Mark is ordered;
4) real profile of extracting is asked to center of gravity;
5) according to 1)～4) step, extract the center of gravity of at least one detection Mark dot profile in addition:
6), after all detection Mark dot profile centers of gravity have been got well in extraction, the profile center of gravity that the center of gravity that benchmark Mark is ordered and detection Mark are ordered compares, and utilizes side-play amount relatively to carry out pcb board compensation.
Step 2) described in image binaryzation processing, exactly the gray-scale value of the pixel on image is only set to 0 or 255, a Gray-scale value is set in tin soldering robot software for calculation, in the time that the gray-scale value of the pixel in the Mark dot image intercepting is more than or equal to Gray-scale value, these pixels are judged as Mark point pixel, and its gray-scale value represents with 255; In the time that the gray-scale value of the pixel in the Mark dot image intercepting is less than Gray-scale value, these pixels are judged as the pixel beyond Mark point, and gray-scale value represents with 0.
Step 3) described in binary image is extracted to the Mark real profile of ordering, the method of extracting has: according to the sum of pixel in Mark point region, obtain the area of profile, obtain the radius of profile according to area,, girth, rectangular degree, circularity, excentricity, breadth length ratio, characterize profile with these.
Step 4) in real profile that Mark is ordered ask the method for center of gravity to be: by profile X, Y coordinate a little add up respectively and average, average coordinates value is center of gravity.
Step 6) in utilize the center of gravity that benchmark Mark orders and detect skew between the center of gravity that Mark order and carry out the method for pcb board compensation and be: the position (can get the comparison of multiple point) of ordering according to the Mark of the Mark point of benchmark and detection, obtain horizontal vertical skew and the anglec of rotation taking benchmark Mark point as reference point, and obtain the rotation offset of pcb board according to this anglec of rotation, and then except carrying out, level, vertical shift compensation, can also being rotated compensation to pcb board to pcb board.
The present invention is owing to taking above technical scheme, and it has the following advantages:
1, this method is by extracting the profile center of gravity that on pcb board, Mark is ordered, and utilizes the skew between benchmark Mark point and detection Mark point center of gravity to carry out pcb board compensation.The shape that algorithm is ordered with Mark is irrelevant, and be not subject to the impact of illumination power, precision is in sub-pix rank, compensation precision is higher, just the precision of compensation has relation with the industrial camera using, when the pixel dimension of camera more hour, the precision of compensation is higher, in the time that the pixel dimension of camera is larger, the precision of compensation is lower.
2, in this method, intercept two even multiple Mark points and calculate center of gravity separately, then compare with benchmark Mark point, multiple points can calculate the anglec of rotation of pcb board accurately, thereby can accurately compensate.If only demand one, can only determine the side-play amount of " X-axis " horizontal direction and " Y-axis " vertical direction.
3, the advantage of binaryzation is to obtain image outline clearly, so this method utilizes binaryzation method to process truncated picture, utilizes acquisition profile to extract the center of gravity of this profile; Ask the method for center of gravity also fairly simple, exactly adding up a little of profile is averaging to coordinate figure.
4, in binary conversion treatment result, the parameter that can characterize real profile has: the rectangular degree of the area of profile, the girth of profile, profile, the circularity of profile, the excentricity of profile and the breadth length ratio of profile, can find the profile of really wanting by these conditions.
, while welding after compensating, can there is not skew and weld inaccurate because the position of pcb board in the pcb board 5, compensating by centre-of gravity shift.
Brief description of the drawings
Fig. 1 is the image (gray-scale map) that intercepts Mark point region with camera;
Fig. 2 is the image (artwork master) after binary conversion treatment.
Embodiment
Below in conjunction with accompanying drawing, the present invention program is described in detail.
The present invention adopts profile gravity model appoach location pcb board, so-called profile gravity model appoach is exactly to take the lead in extracting the profile center of gravity that drafting board Mark to be checked is ordered, and utilizes the skew between benchmark Mark point (benchmark Mark point refers to the Mark point that we extract in programming) and detection Mark point center of gravity to carry out pcb board compensation.Concrete operation step is as follows:
1, with containing the PCB image that Mark is ordered under camera shooting, be presented on computer display screen and (be called FOV visual field, FOV refers to the current captured actual area of camera), then the sectional drawing control carrying with tin soldering robot intercepts and contains the area image that Mark is ordered in current visual field, as shown in Figure 1.
2, tin soldering robot carries out binary conversion treatment to the Mark point area image intercepting, and obtains binary image, i.e. black and white design sketch, as shown in Figure 2.
The binary conversion treatment of image, exactly the gray-scale value of the pixel on image is only set to 0 or 255, a Gray-scale value is set in the software for calculation carrying at tin soldering robot, in the time that the gray-scale value of the pixel in the Mark dot image intercepting is more than or equal to Gray-scale value, these pixels are judged as and belong to certain objects, its gray-scale value represents with 255, is set to white by these pixels; In the time that the gray-scale value of the pixel in the Mark dot image intercepting is less than Gray-scale value, these pixels are excluded beyond object area, and gray-scale value is 0, is set to black by these pixels, represent the object area of background or exception.
3, binary image is extracted to the real profile that Mark is ordered.The real profile of extracting refers to: that annulus of white pixel point and black pixel point intersection in Fig. 2.
In binary conversion treatment result, the parameter that can characterize real profile comprises: the rectangular degree of the area of profile, the girth of profile, profile, the circularity of profile, the excentricity of profile and the breadth length ratio of profile, find us really to want the profile finding by these conditions.Note: in the figure of binary conversion treatment, the region of white pixel point is Mark point region, the region of black pixel point is non-Mark point region.According to the sum of pixel in Mark point region, can obtain the area of profile.After obtaining the area of profile, just can obtain the radius of profile, know that radius just can obtain the rectangular degree of the girth of profile, profile, circularity, the excentricity of profile and the breadth length ratio of profile of profile.
4, according to user's request, select above-mentioned a certain profile characteristic manner, as: select the area of profile, filter out some unwanted profiles by the area of profile, extract interested profile thereby can realize, and extract the center of gravity of this profile.
The real profile that Mark is ordered asks the method for center of gravity to be: adding up a little of profile is averaging to coordinate figure, and average coordinates value is center of gravity.
5,, according to the step of 1-4, obtain the center of gravity of another one Mark dot profile and (ask the benefit of the profile center of gravity that two Mark order to be: can compensate the rotation of pcb board; If only demand a Mark point, can only determine the side-play amount of " X-axis " horizontal direction and " Y-axis " vertical direction.That is to say that to ask a Mark point good, but can affect compensation precision, cause welding inaccurate, institute is in the hope of two Mark points).
6,, after extracting and having got well center of gravity, utilize the skew between the center of gravity that center of gravity that benchmark Mark order and detection Mark order to carry out pcb board compensation.
The method of compensation is: the position of ordering according to the Mark of the Mark point of two benchmark and two detections (or multiple Mark point also can), obtain horizontal vertical skew and the anglec of rotation taking benchmark Mark point as reference point, and obtain the rotation offset of pcb board according to this anglec of rotation, and then except carrying out, level, vertical shift compensation, can also being rotated compensation to pcb board to pcb board.
Emphasis of the present invention is the mark point that finds drafting board to be checked, then compares with benchmark mark point, calculates relatively afterwards side-play amount.The shape that algorithm is ordered with Mark is irrelevant, and precision is in sub-pix rank, and compensation precision is higher.Just the precision of compensation has relation with the industrial camera using, when the pixel dimension of camera more hour, the precision of compensation is higher, in the time that the pixel dimension of camera is larger, the precision of compensation is lower.
Claims (2)
1. a method of ordering with profile gravity model appoach location pcb board Mark, it is characterized in that: the accurate location that is applicable to the front drafting board to be checked of welding, that Mark point to extract when the initial programming is as reference point, adopt profile gravity model appoach by the Mark point contouring center of gravity of drafting board to be checked, the point of the Mark as the reference point center of gravity of extracting during with initial programming is compared, utilize the skew between drafting board Mark point to be checked and benchmark Mark point center of gravity to carry out pcb board compensation, wherein adopt profile gravity model appoach to ask for the method for Mark point center of gravity to be measured as follows: 1) to take lower pcb board to be detected with camera and contain the image that Mark is ordered, then the sectional drawing control carrying with tin soldering robot intercepts and contains the area image that this Mark is ordered in current visual field,
2) the Mark point area image intercepting is carried out to binary conversion treatment, obtain binary image;
3) in binary image, according to the sum of pixel in Mark point area image, obtain the area of image, then obtain the radius of profile according to area, ask for girth, rectangular degree, circularity, excentricity, breadth length ratio according to radius again, with arbitrary parameter characterization Mark dot profile of area, girth, rectangular degree, circularity, excentricity, breadth length ratio;
4) adding up a little of profile is averaging to coordinate figure, average coordinates value is profile center of gravity;
At least choose two and detect Mark point, ask for profile center of gravity separately;
The center of gravity that the profile center of gravity of then ordering with detection Mark and benchmark Mark are ordered compares, and utilizes side-play amount to carry out pcb board compensation; The method of compensation is as follows:
The coordinate of ordering according to benchmark Mark point and the Mark to be detected of two groups of correspondence positions, taking the benchmark Mark point of a group wherein as reference point, obtain the detection Mark point of this correspondence position with respect to level, the vertical shift of this reference point, compensate as level, the vertical offset of pcb board; Taking two benchmark Mark point lines as datum line, obtain line that two Mark to be detected the order anglec of rotation with respect to datum line, compensate as the rotation offset amount of pcb board.
2. method of ordering with profile gravity model appoach location pcb board Mark according to claim 1, it is characterized in that: step 2) described in image binaryzation processing, exactly the gray-scale value of the pixel on image is only set to 0 or 255, a Gray-scale value is set in tin soldering robot software for calculation, in the time that the gray-scale value of the pixel in the Mark dot image intercepting is more than or equal to Gray-scale value, these pixels are judged as Mark point pixel, and its gray-scale value represents with 255; In the time that the gray-scale value of the pixel in the Mark dot image intercepting is less than Gray-scale value, these pixels are judged as the pixel beyond Mark point, and gray-scale value represents with 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210369187.3A CN102930266B (en) | 2012-09-28 | 2012-09-28 | Method for locating Mark points on PCB (printed circuit board) by utilizing outline gravity center method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210369187.3A CN102930266B (en) | 2012-09-28 | 2012-09-28 | Method for locating Mark points on PCB (printed circuit board) by utilizing outline gravity center method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102930266A CN102930266A (en) | 2013-02-13 |
| CN102930266B true CN102930266B (en) | 2014-11-26 |
Family
ID=47645063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210369187.3A Active CN102930266B (en) | 2012-09-28 | 2012-09-28 | Method for locating Mark points on PCB (printed circuit board) by utilizing outline gravity center method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102930266B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466951B (en) * | 2014-09-12 | 2018-11-16 | 江苏明富自动化科技股份有限公司 | A kind of automatic optical detection device and its detection method |
| CN104732207B (en) * | 2015-03-12 | 2018-11-02 | 广东工业大学 | High-precision positions the method and device of pcb board Mark points with high noise immunity |
| CN105427278B (en) * | 2015-10-29 | 2018-03-23 | 广州视源电子科技股份有限公司 | PCB positioning point determining method and system |
| CN107525477B (en) * | 2016-06-22 | 2019-09-20 | 大族激光科技产业集团股份有限公司 | A kind of 5 points of identification CCD methods |
| CN107578431A (en) * | 2017-07-31 | 2018-01-12 | 深圳市海思科自动化技术有限公司 | A kind of Mark points visual identity method |
| CN107967679B (en) * | 2017-11-21 | 2020-07-03 | 凌云光技术集团有限责任公司 | Method for automatically selecting positioning core based on vector graph of PCB product |
| CN108363508B (en) * | 2018-01-13 | 2021-03-23 | 江南大学 | Mark positioning non-contact visual detection method for mobile phone touch screen |
| CN108447089A (en) * | 2018-03-27 | 2018-08-24 | 凌云光技术集团有限责任公司 | A kind of PCB figures major axes orientation computational methods |
| CN109064481B (en) * | 2018-07-24 | 2021-09-17 | 中山新诺科技股份有限公司 | Machine vision positioning method |
| CN109660697B (en) * | 2018-12-19 | 2021-06-18 | 森大（深圳）技术有限公司 | Method, device and equipment for automatically correcting positioning error of PCB (printed Circuit Board) and storage medium |
| CN109685784A (en) * | 2018-12-19 | 2019-04-26 | 森大（深圳）技术有限公司 | Pcb board position error calibration method, device, equipment and storage medium |
| CN109785324B (en) * | 2019-02-01 | 2020-11-27 | 佛山市南海区广工大数控装备协同创新研究院 | Large-format PCB positioning method |
| CN109859198B (en) * | 2019-02-01 | 2020-10-30 | 佛山市南海区广工大数控装备协同创新研究院 | Large-breadth PCB multi-region accurate positioning method |
| CN110587992B (en) * | 2019-09-05 | 2022-01-25 | 深圳市巨力方视觉技术有限公司 | Article attaching method, device and computer-readable storage medium |
| CN117372434B (en) * | 2023-12-08 | 2024-04-30 | 深圳市强达电路股份有限公司 | Positioning system and method for PCB production |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101358836A (en) * | 2008-09-28 | 2009-02-04 | 西安理工大学 | Method for recognising weld spot central position based on computer vision |
-
2012
- 2012-09-28 CN CN201210369187.3A patent/CN102930266B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101358836A (en) * | 2008-09-28 | 2009-02-04 | 西安理工大学 | Method for recognising weld spot central position based on computer vision |
Non-Patent Citations (5)
| Title |
|---|
| Hough变换快速检测圆形PCB Mark;柏长冰等;《光电工程》;20050930;第32卷(第9期);75－78 * |
| PCB检测中圆Mark的快速精确定位;熊光洁等;《计算机工程》;20120630;第38卷(第11期);211-213 * |
| 李程鹏等.基于HALCON的PCB光学定位点的3种识别方法及比较.《佛山科学技术学院学报（自然科学版）》.2010,第28卷(第2期),29-33. * |
| 柏长冰等.Hough变换快速检测圆形PCB Mark.《光电工程》.2005,第32卷(第9期),75－78. * |
| 熊光洁等.PCB检测中圆Mark的快速精确定位.《计算机工程》.2012,第38卷(第11期),211-213. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102930266A (en) | 2013-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102930266B (en) | Method for locating Mark points on PCB (printed circuit board) by utilizing outline gravity center method | |
| CN102873420B (en) | Method for positioning Mark points of PCB (printed circuit board) by image matching | |
| CN106546263B (en) | A kind of laser leveler shoot laser line detecting method based on machine vision | |
| CN108363508B (en) | Mark positioning non-contact visual detection method for mobile phone touch screen | |
| CN102878941B (en) | Method for positioning Mark points of PCB (printed circuit board) by circular profile method | |
| CN110501342B (en) | Cheese yarn rod positioning visual detection method | |
| CN113240674B (en) | Coplanarity detection method based on three-dimensional point cloud and two-dimensional image fusion | |
| CN105783711B (en) | Three-dimensional scanner correction system and correction method thereof | |
| CN110213900B (en) | PCB printing machine vision detection method and system based on CCD recognition | |
| CN113724193B (en) | PCBA part size and clearance high-precision visual measurement method | |
| CN114220757B (en) | Wafer detection alignment method, device and system and computer medium | |
| CN113269762B (en) | Screen defect detection method, system and computer storage medium | |
| CN106501272B (en) | Machine vision soldering tin positioning detection system | |
| CN104331695B (en) | A kind of circle marker symbol shape quality detection method of robust | |
| CN108871185B (en) | Method, device and equipment for detecting parts and computer readable storage medium | |
| US20190080468A1 (en) | Positioning and measuring system based on image scale | |
| CN113538583A (en) | Method for accurately positioning position of workpiece on machine tool and vision system | |
| CN108709500B (en) | Circuit board element positioning and matching method | |
| CN110695520A (en) | Vision-based full-automatic galvanometer field calibration system and calibration method thereof | |
| CN112419225A (en) | SOP type chip detection method and system based on pin segmentation | |
| Wu et al. | Accurate localization of defective circular PCB mark based on sub-pixel edge detection and least square fitting | |
| CN105431032A (en) | Substrate detection method | |
| US11357149B2 (en) | Component orientation determination data creation device and component orientation determination data creation method | |
| CN104677906A (en) | Image information detecting method | |
| JP2008185395A (en) | Mounting substrate visual inspection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |