CN116612071A - Accurate positioning method for large-format PCB panel with repeated elements - Google Patents
Accurate positioning method for large-format PCB panel with repeated elements Download PDFInfo
- Publication number
- CN116612071A CN116612071A CN202310418227.7A CN202310418227A CN116612071A CN 116612071 A CN116612071 A CN 116612071A CN 202310418227 A CN202310418227 A CN 202310418227A CN 116612071 A CN116612071 A CN 116612071A
- Authority
- CN
- China
- Prior art keywords
- points
- coordinates
- mark
- image data
- initial
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 82
- 238000001514 detection method Methods 0.000 claims abstract description 41
- 239000011159 matrix material Substances 0.000 claims description 75
- 238000001914 filtration Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 238000013519 translation Methods 0.000 claims description 9
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000007547 defect Effects 0.000 abstract description 11
- 238000013135 deep learning Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011810 insulating material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003252 repetitive effect Effects 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000011889 copper foil Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000945 filler Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010030 laminating Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30148—Semiconductor; IC; Wafer
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides a large-format PCB panel accurate positioning method with repeated elements, which comprises nine-point calibration, initializing industrial camera parameters, carrying out coarse positioning of Mark points by pyramid downsampling and Hough circle detection, carrying out positioning of two Mark points at the upper left and lower right by combining fine positioning of sub-pixel level by using a caliper circle detection algorithm, converting coordinates of a point set under a DXF coordinate system into pulse coordinates, converting photographing position information into pixel coordinates to obtain pixel coordinates of a component coordinate point set, and carrying out de-duplication on coordinate information of the repeated components adjacent to the left and right of the FOV by a KNN algorithm, thereby realizing high-accuracy, high-precision and strong-adaptability rapid positioning detection, simultaneously being capable of carrying out accurate positioning and detection on each component on different PCB products rapidly and accurately, and greatly improving the defect detection efficiency of electronic components in the PCB panel.
Description
Technical Field
The invention relates to the technical field of machine vision and defect detection of PCB panel electronic components, in particular to a large-format PCB panel accurate positioning method with repeated elements.
Background
In the field of equipment manufacturing, PCB circuit boards are important electronic components, and there are single-panel, double-panel and multi-layer boards for circuit boards. On the most basic PCB, where the parts are concentrated on one side and the wires on the other side, such a PCB is called a single panel. The double-sided board is provided with copper-clad wiring on both sides, and the wiring between the two layers can be conducted through the via holes so as to form the required network connection. The multi-layer board is formed by laminating more than three conductive pattern layers with insulating materials therebetween at intervals, and the conductive patterns therebetween are interconnected as required. The PCB circuit board mainly comprises a bonding pad, a through hole, a mounting hole, an electric wire, a component, a connector, a filler, an electric appliance boundary and the like, and the circuit board is separated from the surface copper foil conductive layer by utilizing a board-based insulating material, so that current flows in various components along a pre-designed route to finish functions such as acting, amplifying, attenuating, modulating, demodulating, encoding and the like.
The PCB panel positioning detection mainly faces the problems that electronic components in the PCB panel are arranged tightly in tiny mode, the number of components in a single PCB panel is huge, imaging requirements on a PCB are strict, the whole product of the PCB panel is divided into a plurality of fields of view after imaging, the components in different fields of view can be repeated or lack, and positioning detection robustness on different PCB products is poor.
In the current technology, the conventional positioning method of the PCB product is mostly dependent on manual or mechanical positioning modes. The accuracy of the method depends on the accuracy of the manual work and the mechanical design, and the method can interfere the positioning of the manual naked eyes under the non-uniform illumination condition and has no reliability. On the other hand, the problems of degradation of the shot PCB image structure, low contrast, blurred edges and the like exist by utilizing a visual positioning technology, so that serious deviation of positioning occurs.
The prior art at present discloses a circuit board surface defect detection method based on deep learning, the device comprises a conveying and positioning device, an image acquisition device, an image processing unit, a deep learning analysis processing unit and a deep learning analysis processing unit, wherein the surface image of the printed circuit board preprocessed by the image processing unit is analyzed and processed through a preset deep learning algorithm, and a detection result is given out so as to realize rapid and accurate circuit board surface defect detection; in the prior art, although the neural network is used for analyzing and processing the image of the PCB to improve the detection efficiency, the image acquisition mode is still to carry out shooting acquisition by setting industrial cameras with different angles, and the problems of degradation of the shot image structure of the PCB, low contrast, blurred edges and the like exist, so that the positioning deviation and the detection precision are reduced; in addition, the device faces the circuit board with huge components, the shot image can also have the problems of lower precision, device ghost and the like, each component can not be accurately positioned, and the detection precision is further reduced.
Disclosure of Invention
The invention provides a large-format PCB panel accurate positioning method with repeated elements, which can rapidly and accurately position and detect each component on different PCB products in order to overcome the defects of lower positioning detection precision and repeated positioning or positioning deficiency when facing the PCB with large product size and huge component quantity in the prior art.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a method for accurately positioning a large-format PCB panel with repeated elements comprises the following steps:
s1: acquiring initial pixel coordinates and initial pulse coordinates of Mark points on a large-format PCB panel with repeated elements by using an industrial camera and a calibration plate through a nine-point calibration method, and acquiring a homography matrix between a pulse coordinate system and a pixel coordinate system, and initial slopes and initial angles between two Mark points at the upper left and the lower right;
s2: obtaining DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the homography matrix;
s3: preprocessing the PCB image data obtained by shooting at each shooting position, respectively performing coarse positioning and fine positioning on two Mark points at the upper left and lower right in the preprocessed PCB image data, and obtaining pixel coordinates of the two Mark points at the upper left and lower right under a pixel coordinate system;
s4: converting pixel coordinates of the upper left Mark point and the lower right Mark point under a pixel coordinate system into pulse coordinates by utilizing a homography matrix, calculating the slope and the angle between the upper left Mark point and the lower right Mark point again, and rotating the preprocessed DXF image data by taking the center of the DXF coordinate system as a rotation center, so that the slope between the upper left Mark point and the lower right Mark point is the same as the initial slope, and completing the alignment of the coordinate system;
s5: according to a preset photographing position, using the initialized industrial camera to perform point set photographing, setting a plurality of fields of view in each photographing position, outputting pulse coordinates of points in the field of view after one field of view is photographed, and removing the points in the field of view from a total point set in DXF image data;
after traversing the total point set in the DXF image data, acquiring pulse coordinates of all points, and converting the pulse coordinates of each point into pixel coordinates by utilizing a homography matrix;
s6: and (3) performing de-duplication on repeated points in the traversal process of the DXF image data total point set by using a KNN algorithm, and storing and outputting pixel coordinates of all the de-duplicated points to finish accurate positioning of all elements.
Preferably, in the step S1, an industrial camera and a calibration board are used to obtain an initial pixel coordinate and an initial pulse coordinate of a Mark point on a large-format PCB panel with a repetitive element under a pulse coordinate system by a nine-point calibration method, and a homography matrix between the initial pulse coordinate and the initial pixel coordinate is obtained, and the specific method is as follows:
fixing an industrial camera on AOI industrial detection equipment, wherein the height of the industrial camera is the same as that of a large-format PCB panel with repeated elements; the calibration plate is arranged below the camera, and the position area is consistent with the working area of the industrial camera scanning PCB panel of the machine;
setting the focal length of an industrial camera, photographing, carrying out gray scale processing on the photographed image, identifying initial pixel coordinates of all Mark points by adopting a contour rounding method, generating initial pulse coordinates of all Mark points on a calibration plate according to the distance between Mark points input into the calibration plate and the position information of the first Mark point, calculating and obtaining a homography matrix according to the obtained initial pixel coordinates of the Mark points and the initial pulse coordinates corresponding to the Mark points, and recording the pulse coordinates of the current photographing position.
Preferably, the homography matrix includes a rotation matrix and a translation matrix, specifically:
wherein, (x) c ,y c ) For the initial pixel coordinates, (x) t ,y t ) For the initial pulse coordinates, R is a rotation matrix, which satisfiesM is a translation matrix, satisfying->
According to the obtained initial pixel coordinates of Mark points and the corresponding initial pulse coordinates, calculating a homography matrix by using the following formula:
wherein, (x) c1 ,y c1 ),(x c2 ,y c2 ),(x c3 ,y c3 ) Initial pixel coordinates of the first, second and third Mark, respectively; (x) t1 ,y t1 ),(x t2 ,y t2 ),(x t3 ,y t3 ) Initial pulse coordinates of the first, second and third Mark, respectively;
and equally dividing the initial pixel coordinates and the initial pulse coordinates of the nine Mark points into three groups, calculating three times according to the formula, and taking the average value as a final homography matrix.
Preferably, in the step S2, the specific method for inputting the homography matrix into the industrial camera and initializing all the parameters thereof is as follows:
the homography matrix is entered into the industrial camera and initializes the focal length, principal distance, distortion coefficient, adjacent pixel distance of the industrial camera, and the actual range of the single field of view in the DXF image data.
Preferably, in the step S3, the specific method for preprocessing the PCB image data obtained by shooting at each shooting position is as follows:
and carrying out graying treatment on the PCB image data obtained by shooting at each shooting position, and carrying out Gaussian filtering treatment and median filtering treatment on the PCB image data subjected to the graying treatment to complete pretreatment.
Preferably, in the step S3, the specific method for coarsely positioning the two Mark points on the upper left and the lower right in the preprocessed PCB image data includes:
performing pyramid downsampling processing for 3 times on surrounding areas of upper left Mark points and lower right Mark points in the preprocessed PCB image data;
according to Hough circle detection, the first stage is used for detecting circle centers of upper left Mark points and lower right Mark points, the second stage derives circle radiuses from the circle centers, and coarse positioning of the upper left Mark points and the lower right Mark points is completed through the two stages.
Preferably, in the step S3, the specific method for precisely positioning the two Mark points on the upper left and the lower right in the preprocessed PCB image data includes:
and (3) carrying out Gaussian filtering on the outlines of the two Mark points at the upper left and the lower right after coarse positioning by using a caliper rule detection circle method, extracting edge points, finishing fine positioning, and obtaining pixel coordinates of the two Mark points at the upper left and the lower right under a pixel coordinate system.
Preferably, the positioning accuracy of the fine positioning is at the sub-pixel level.
Preferably, in step S4, the preprocessed DXF image data is rotated with the DXF coordinate system center as a rotation center, specifically:
the initial slope and the initial angle of the upper left Mark point and the lower right Mark point are respectively recorded as k 0 And a 0 The calculated slope and angle are respectively recorded as k 1 And a 1 ；
Rotating the preprocessed DXF image data by taking the center of a DXF coordinate system as a rotation center, wherein the rotation angle is a 1 -a 0 Such that the slope k between the upper left and lower right Mark points and the initial slope k 0 The same is done for the alignment of the coordinate system.
Preferably, in the step S6, the specific method for performing deduplication on the repeated points in the traversal process of the DXF image data total point set by using the KNN algorithm is as follows:
calculating the distance between a point and the pixel coordinates of all other points by utilizing a hamming distance formula, and sequencing all the distances from small to large;
selecting the first K points with the smallest distance, inputting the first K points into a trained KNN model, respectively calculating the similarity between the first K points and the K points, and rejecting the points with the similarity higher than a preset threshold value as repeated points;
and performing de-duplication treatment on repeated points in all points by using the trained KNN model and a hamming distance formula.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the invention provides a large-format PCB panel accurate positioning method with a repeating element, which comprises the steps of obtaining initial pixel coordinates and initial pulse coordinates of Mark points on the large-format PCB panel with the repeating element by using an industrial camera and a calibration plate through a nine-point calibration method, and obtaining a homography matrix between the pulse coordinates and the pixel coordinates, and initial slopes and initial angles between two Mark points on the upper left and the lower right; obtaining DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the homography matrix; preprocessing the PCB image data obtained by shooting at each shooting position, respectively performing coarse positioning and fine positioning on two Mark points at the upper left and lower right in the preprocessed PCB image data, and obtaining pixel coordinates of the two Mark points at the upper left and lower right under a pixel coordinate system; converting pixel coordinates of the upper left Mark point and the lower right Mark point under a pixel coordinate system into pulse coordinates by utilizing a homography matrix, calculating the slope and the angle between the upper left Mark point and the lower right Mark point again, and rotating the preprocessed DXF image data by taking the center of the DXF coordinate system as a rotation center, so that the slope between the upper left Mark point and the lower right Mark point is the same as the initial slope, and completing the alignment of the coordinate system; according to a preset photographing position, using the initialized industrial camera to perform point set photographing, setting a plurality of fields of view in each photographing position, outputting pulse coordinates of points in the field of view after one field of view is photographed, and removing the points in the field of view from a total point set in DXF image data; after traversing the total point set in the DXF image data, acquiring pulse coordinates of all points, and converting the pulse coordinates of each point into pixel coordinates by utilizing a homography matrix; repeating points in the traversal process of the DXF image data total point set are de-duplicated by using a KNN algorithm, pixel coordinates of all the de-duplicated points are stored and output, and accurate positioning of all elements is completed;
aiming at the large-breadth PCB panel with repeated and large number of tiny components, the positioning method provided by the invention can effectively avoid the defects of inaccurate positioning, poor precision and poor adaptability of the traditional machine vision positioning algorithm; according to the invention, coarse positioning of Mark points is carried out by pyramid downsampling and Hough circle detection, fine positioning of sub-pixel level is carried out by using a calliper circle detection algorithm, coordinates of a point set under a DXF coordinate system are converted into pulse coordinates, photographing position information is combined and converted into pixel coordinates to obtain pixel coordinates of a component coordinate set, and repeated component coordinate information adjacent to the left and right of the FOV is de-duplicated by a KNN algorithm, so that high-accuracy, high-precision and strong-adaptability rapid positioning detection is realized, and meanwhile, accurate positioning and detection can be carried out on each component on different PCB products rapidly and accurately, and the defect detection efficiency of electronic components in a PCB panel is greatly improved.
Drawings
Fig. 1 is a flowchart of a method for precisely positioning a large-format PCB panel with repeating elements according to embodiment 1.
Fig. 2 is a flowchart of a method for precisely positioning a large-format PCB panel with repeating elements according to embodiment 2.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the present patent;
for the purpose of better illustrating the embodiments, certain elements of the drawings may be omitted, enlarged or reduced and do not represent the actual product dimensions;
it will be appreciated by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical scheme of the invention is further described below with reference to the accompanying drawings and examples.
Example 1
As shown in fig. 1, the embodiment provides a method for precisely positioning a large-format PCB panel with a repeating element, including the following steps:
s1: acquiring initial pixel coordinates and initial pulse coordinates of Mark points on a large-format PCB panel with repeated elements by using an industrial camera and a calibration plate through a nine-point calibration method, and acquiring a homography matrix between a pulse coordinate system and a pixel coordinate system, and initial slopes and initial angles between two Mark points at the upper left and the lower right;
s2: obtaining DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the homography matrix;
s3: preprocessing the PCB image data obtained by shooting at each shooting position, respectively performing coarse positioning and fine positioning on two Mark points at the upper left and lower right in the preprocessed PCB image data, and obtaining pixel coordinates of the two Mark points at the upper left and lower right under a pixel coordinate system;
s4: converting pixel coordinates of the upper left Mark point and the lower right Mark point under a pixel coordinate system into pulse coordinates by utilizing a homography matrix, calculating the slope and the angle between the upper left Mark point and the lower right Mark point again, and rotating the preprocessed DXF image data by taking the center of the DXF coordinate system as a rotation center, so that the slope between the upper left Mark point and the lower right Mark point is the same as the initial slope, and completing the alignment of the coordinate system;
s5: according to a preset photographing position, using the initialized industrial camera to perform point set photographing, setting a plurality of fields of view in each photographing position, outputting pulse coordinates of points in the field of view after one field of view is photographed, and removing the points in the field of view from a total point set in DXF image data;
after traversing the total point set in the DXF image data, acquiring pulse coordinates of all points, and converting the pulse coordinates of each point into pixel coordinates by utilizing a homography matrix;
s6: and (3) performing de-duplication on repeated points in the traversal process of the DXF image data total point set by using a KNN algorithm, and storing and outputting pixel coordinates of all the de-duplicated points to finish accurate positioning of all elements.
In the specific implementation process, firstly, an industrial camera and a calibration plate are utilized to obtain initial pixel coordinates and initial pulse coordinates of Mark points on a large-format PCB panel with repeated elements through a nine-point calibration method, and a homography matrix between a pulse coordinate system and the pixel coordinate system, and initial slopes and initial angles between two Mark points at the upper left and the lower right are obtained; obtaining DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the homography matrix; preprocessing the PCB image data obtained by shooting at each shooting position, respectively performing coarse positioning and fine positioning on two Mark points at the upper left and lower right in the preprocessed PCB image data, and obtaining pixel coordinates of the two Mark points at the upper left and lower right under a pixel coordinate system; converting pixel coordinates of the upper left Mark point and the lower right Mark point under a pixel coordinate system into pulse coordinates by utilizing a homography matrix, calculating the slope and the angle between the upper left Mark point and the lower right Mark point again, and rotating the preprocessed DXF image data by taking the center of the DXF coordinate system as a rotation center, so that the slope between the upper left Mark point and the lower right Mark point is the same as the initial slope, and completing the alignment of the coordinate system; according to a preset photographing position, using the initialized industrial camera to perform point set photographing, setting a plurality of fields of view in each photographing position, outputting pulse coordinates of points in the field of view after one field of view is photographed, and removing the points in the field of view from a total point set in DXF image data; after traversing the total point set in the DXF image data, acquiring pulse coordinates of all points, and converting the pulse coordinates of each point into pixel coordinates by utilizing a homography matrix; finally, repeating points in the traversal process of the DXF image data total point set are subjected to de-duplication by utilizing a KNN algorithm, pixel coordinates of all the points after de-duplication are stored and output, and accurate positioning of all elements is completed;
the method aims at the large-format PCB panel with repeated and large number of tiny components, and the defects of inaccurate positioning, poor precision and poor adaptability of the traditional machine vision positioning algorithm can be effectively avoided; according to the method, coarse positioning of Mark points is carried out by pyramid downsampling and Hough circle detection, fine positioning of sub-pixel levels is carried out by using a calliper circle detection algorithm, coordinates of a point set under a DXF coordinate system are converted into pulse coordinates, photographing position information is combined and converted into pixel coordinates to obtain pixel coordinates of a component coordinate set, repeated component coordinate information adjacent to the left side and the right side of the FOV is subjected to deduplication through a KNN algorithm, and therefore high-accuracy, high-precision and strong-adaptability rapid positioning detection is achieved, meanwhile, accurate positioning and detection can be carried out on each component on different PCB products rapidly and accurately, and the defect detection efficiency of electronic components in a PCB panel is greatly improved.
Example 2
As shown in fig. 2, the embodiment provides a method for precisely positioning a large-format PCB panel with a repeating element, which includes the following steps:
s1: acquiring initial pixel coordinates and initial pulse coordinates of Mark points on a large-format PCB panel with repeated elements by using an industrial camera and a calibration plate through a nine-point calibration method, and acquiring a homography matrix between a pulse coordinate system and a pixel coordinate system, and initial slopes and initial angles between two Mark points at the upper left and the lower right;
s2: obtaining DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the homography matrix;
s3: preprocessing the PCB image data obtained by shooting at each shooting position, respectively performing coarse positioning and fine positioning on two Mark points at the upper left and lower right in the preprocessed PCB image data, and obtaining pixel coordinates of the two Mark points at the upper left and lower right under a pixel coordinate system;
s4: converting pixel coordinates of the upper left Mark point and the lower right Mark point under a pixel coordinate system into pulse coordinates by utilizing a homography matrix, calculating the slope and the angle between the upper left Mark point and the lower right Mark point again, and rotating the preprocessed DXF image data by taking the center of the DXF coordinate system as a rotation center, so that the slope between the upper left Mark point and the lower right Mark point is the same as the initial slope, and completing the alignment of the coordinate system;
s5: according to a preset photographing position, using the initialized industrial camera to perform point set photographing, setting a plurality of fields of view in each photographing position, outputting pulse coordinates of points in the field of view after one field of view is photographed, and removing the points in the field of view from a total point set in DXF image data;
after traversing the total point set in the DXF image data, acquiring pulse coordinates of all points, and converting the pulse coordinates of each point into pixel coordinates by utilizing a homography matrix;
s6: repeating points in the traversal process of the DXF image data total point set are de-duplicated by using a KNN algorithm, pixel coordinates of all the de-duplicated points are stored and output, and accurate positioning of all elements is completed;
in the step S1, an industrial camera and a calibration board are used to obtain an initial pixel coordinate and an initial pulse coordinate of Mark points on a large-format PCB panel with a repetitive element under a pulse coordinate system by a nine-point calibration method, and a homography matrix between the initial pulse coordinate and the initial pixel coordinate is obtained, and the specific method is as follows:
fixing an industrial camera on AOI industrial detection equipment, wherein the height of the industrial camera is the same as that of a large-format PCB panel with repeated elements; the calibration plate is arranged below the camera, and the position area is consistent with the working area of the industrial camera scanning PCB panel of the machine;
setting the focal length of an industrial camera, photographing, carrying out gray scale processing on the photographed image, identifying initial pixel coordinates of all Mark points by adopting a contour rounding method, generating initial pulse coordinates of all Mark points on a calibration plate according to the distance between Mark points of the input calibration plate and the position information of a first Mark point, calculating and obtaining a homography matrix according to the obtained initial pixel coordinates of the Mark points and the initial pulse coordinates corresponding to the Mark points, and recording the pulse coordinates of the current photographing position;
the homography matrix comprises a rotation matrix and a translation matrix, and specifically comprises the following steps:
wherein, (x) c ,y c ) For the initial pixel coordinates, (x) t ,y t ) For the initial pulse coordinates, R is a rotation matrix, which satisfiesM is a translation matrix, satisfying->
According to the obtained initial pixel coordinates of Mark points and the corresponding initial pulse coordinates, calculating a homography matrix by using the following formula:
wherein, (x) c1 ,y c1 ),(x c2 ,y c2 ),(x c3 ,y c3 ) Initial pixel coordinates of the first, second and third Mark, respectively; (x) t1 ,y t1 ),(x t2 ,y t2 ),(x t3 ,y t3 ) Initial pulse coordinates of the first, second and third Mark, respectively;
dividing the initial pixel coordinates and the initial pulse coordinates of the nine Mark points into three groups, calculating three times according to the formula, and taking the average value as a final homography matrix;
in the step S2, the specific method for inputting the homography matrix into the industrial camera and initializing all the parameters thereof is as follows:
inputting a homography matrix into the industrial camera and initializing the focal length, main distance, distortion coefficient, adjacent pixel distance of the industrial camera, and the actual range of a single field of view in the DXF image data;
in the step S3, the specific method for preprocessing the PCB image data obtained by shooting at each shooting position includes:
preprocessing the PCB image data obtained by shooting at each shooting position to carry out graying treatment, and carrying out Gaussian filtering treatment and median filtering treatment on the PCB image data subjected to the graying treatment to finish preprocessing;
in the step S3, the specific method for coarsely positioning the two Mark points on the upper left and the lower right in the preprocessed PCB image data includes:
performing pyramid downsampling processing for 3 times on surrounding areas of upper left Mark points and lower right Mark points in the preprocessed PCB image data;
according to Hough circle detection, the first stage is used for detecting circle centers of upper left Mark points and lower right Mark points, the second stage derives circle radii from the circle centers, and coarse positioning of the upper left Mark points and the lower right Mark points is completed through the two stages;
in the step S3, the specific method for precisely positioning the two Mark points on the upper left and the lower right in the preprocessed PCB image data includes:
carrying out Gaussian filtering on the outlines of the two Mark points at the upper left and the lower right after coarse positioning by using a caliper rule detection circle method, extracting edge points, finishing fine positioning, and obtaining pixel coordinates of the two Mark points at the upper left and the lower right under pixels;
the positioning precision of the fine positioning is at a sub-pixel level;
in the step S4, the preprocessed DXF image data is rotated with the DXF coordinate system center as the rotation center, specifically:
the initial slope and the initial angle of the upper left Mark point and the lower right Mark point are respectively recorded as k 0 And a 0 The calculated slope and angle are respectively recorded as k 1 And a 1 ；
Rotating the preprocessed DXF image data by taking the center of a DXF coordinate system as a rotation center, wherein the rotation angle is a 1 -a 0 Such that the slope k between the upper left and lower right Mark points and the initial slope k 0 The same, finish the alignment of the coordinate system;
in the step S6, the specific method for performing deduplication on the repeated points in the traversal process of the DXF image data total point set by using the KNN algorithm is as follows:
calculating the distance between a point and the pixel coordinates of all other points by utilizing a hamming distance formula, and sequencing all the distances from small to large;
selecting the first K points with the smallest distance, inputting the first K points into a trained KNN model, respectively calculating the similarity between the first K points and the K points, and rejecting the points with the similarity higher than a preset threshold value as repeated points;
and performing de-duplication treatment on repeated points in all points by using the trained KNN model and a hamming distance formula.
In the specific implementation process, in order to determine the correlation between the three-dimensional geometric position of a certain point on the surface of the PCB panel and the corresponding point in the image, the panel is required to be calibrated, and the calibration is favorable for high-precision measurement and positioning; the space object presents a three-dimensional geometric position, and a projection image in the camera is a two-dimensional position, so that the calibration is to determine the relationship between the three-dimensional geometric position of a certain point of the space object and a corresponding point in the projection image; calibration must establish a projection mathematical model of the object and the camera, i.e. a geometric model of camera imaging; the mathematical parameters for constructing the geometric model are the contents to be calibrated, namely the internal and external parameters of the camera, the internal parameters are the technological parameters of the camera, including focal length, main distance, distortion coefficient, adjacent pixel distance and the like, and the external parameters are the position parameters of the camera, including rotation angle, translation distance and the like; in the embodiment, a nine-point calibration method in industrial vision is selected to calibrate the PCB panel, so that a conversion relation of converting a pixel coordinate system into a pulse coordinate system is obtained;
firstly, acquiring initial pixel coordinates and initial pulse coordinates of Mark points on a large-format PCB panel with repeated elements by using an industrial camera and a calibration plate through a nine-point calibration method, and acquiring a homography matrix between a pulse coordinate system and a pixel coordinate system, and initial slopes and initial angles between two Mark points at the upper left and the lower right, wherein the method comprises the following steps:
fixing an industrial camera on AOI industrial detection equipment, wherein the height of the industrial camera is the same as that of a large-format PCB panel with repeated elements; the calibration plate is arranged below the camera, and the position area is consistent with the working area of the industrial camera scanning PCB panel of the machine;
setting the focal length of an industrial camera, photographing, carrying out gray scale processing on the photographed image, identifying initial pixel coordinates of all Mark points by adopting a contour rounding method, generating initial pulse coordinates of all Mark points on a calibration plate according to the distance between Mark points of the input calibration plate and the position information of a first Mark point, calculating and obtaining a homography matrix according to the obtained initial pixel coordinates of the Mark points and the initial pulse coordinates corresponding to the Mark points, and recording the pulse coordinates of the current photographing position;
the homography matrix comprises a rotation matrix and a translation matrix, and specifically comprises the following steps:
wherein, (x) c ,y c ) For the initial pixel coordinates, (x) t ,y t ) For the initial pulse coordinates, R is a rotation matrix, which satisfiesM is a translation matrix, satisfying->
According to the obtained initial pixel coordinates of Mark points and the corresponding initial pulse coordinates, calculating a homography matrix by using the following formula:
wherein, (x) c1 ,y c1 ),(x c2 ,y c2 ),(x c3 ,y c3 ) Initial pixel coordinates of the first, second and third Mark, respectively; (x) t1 ,y t1 ),(x t2 ,y t2 ),(x t3 ,y t3 ) Initial pulse coordinates of the first, second and third Mark, respectively;
dividing the initial pixel coordinates and the initial pulse coordinates of the nine Mark points into three groups, calculating three times according to the formula, and taking the average value as a final homography matrix, wherein the position coordinates of the z axis can be set as required to finish calibration;
acquiring DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the industrial camera, wherein the parameters comprise focal length, main distance, distortion coefficient, adjacent pixel distance of the industrial camera and actual range of a single view field in the DXF image data;
and then preprocessing the PCB image data, wherein the specific method comprises the following steps:
carrying out graying treatment on the PCB image data, and carrying out Gaussian filtering treatment and median filtering treatment on the PCB image data after the graying treatment to complete pretreatment;
respectively performing coarse positioning and fine positioning on two Mark points at the upper left and the lower right in the preprocessed PCB image data to obtain pixel coordinates of the two Mark points at the upper left and the lower right in a pixel coordinate system;
because of the oversized PCB panel, photographing imaging cannot be completed at one time and needs to be divided into a plurality of fields of view. Therefore, a plurality of view fields are required to be positioned, and a small area of two Mark points at the upper left corner and the lower right corner on the PCB is imaged to be positioned in a twice positioning mode; in the embodiment, the Mark point is positioned twice, coarse positioning is performed first, and then fine positioning is performed again on the basis, so that the error of Mark point positioning is greatly reduced;
in the embodiment, the coarse positioning adopts an image pyramid to divide Mark point images, gaussian kernel convolution is carried out on the images to enable adjacent pixels to have higher weight, the blurring effect is achieved, then all even rows and columns are deleted, and downsampling is carried out on the images;
in the embodiment, the surrounding areas of the left upper Mark point and the right lower Mark point in the preprocessed PCB image data are subjected to pyramid downsampling for 3 times, coarse positioning is performed according to Hough circle detection, the first stage is used for detecting the circle center, and the second stage derives the circle radius from the circle center; the principle of detecting the circle center is that the circle center is the intersection of all normals of the circle where the circle center is located, so that the circle center can be determined only by finding the intersection, the method of detecting the radius of the circle is that the distance (namely the radius) from the circle center to any point on the circle is the same, only a threshold value is needed to be determined, and the distance can be considered to be the radius of the circle corresponding to the circle center as long as the number of the same distances is larger than the threshold value, and a circle can be obtained after the circle center and the radius are obtained, so that the rough positioning of Mark points is completed;
performing fine positioning on the basis of coarse positioning, wherein the fine positioning is sub-pixel level positioning; the method adopts a caliper to detect circles, and the principle of the method is to find the place where gray values in N small rectangular ROIs are suddenly changed; firstly, through each rectangular ROI of a caliper, calculating an average gray value (only when a contour line and a boundary of an image to be measured are not vertical) along a slicing direction, and performing Gaussian filtering on the average gray value (contour), wherein the purpose of Gaussian filtering is to enable a curve to be smoother and eliminate noise points; extracting edge points according to preset parameters, and extracting sub-pixels; the edge detection with strong anti-interference performance, strong robustness and high precision can be realized through coarse positioning and fine positioning, so that the positioning precision of Mark points is improved;
because there may be inclination of the PCB panel during calibration in step S1, in step S4, DXF image data needs to be aligned with the actual PCB board, the pixel coordinates of the upper left and lower right Mark points in the pixel coordinate system are converted into pulse coordinates by using a homography matrix, and the slope and angle between the upper left and lower right Mark points are calculated again, and the preprocessed DXF image data is rotated with the center of the DXF coordinate system as the rotation center, so that the slope between the upper left and lower right Mark points is the same as the initial slope, and alignment of the coordinate system is completed, specifically:
the initial slope and the initial angle of the upper left Mark point and the lower right Mark point are respectively recorded as k 0 And a 0 The calculated slope and angle are respectively recorded as k 1 And a 1 ；
Rotating the preprocessed DXF image data by taking the center of a DXF coordinate system as a rotation center, wherein the rotation angle is a 1 -a 0 Such that the slope k between the upper left and lower right Mark points and the initial slope k 0 The same, finish the alignment of the coordinate system;
then, shooting a point set by using the initialized industrial camera according to a preset shooting position, setting a plurality of fields of view in each shooting position, outputting pulse coordinates of points in the field of view after each shooting of one field of view, removing the points in the field of view from a total point set in DXF image data, and then continuously shooting the next field of view;
the operation is circulated until the total point set is traversed after all photographing bits are photographed, pulse coordinates of all points are obtained after the total point set in DXF image data is traversed, and the pulse coordinates of each point are converted into pixel coordinates by utilizing a homography matrix;
after the pixel coordinates of all the point sets are obtained, a large-format repeated target point set appears in the adjacent fields of view, because one part of some elements is at one field of view edge, the other part is at the other field of view edge, and the points of the elements selected after the two fields of view are shot are repeated;
therefore, the repeated points in the traversal process of the DXF image data total point set are de-duplicated finally by using the KNN algorithm, and the specific method is as follows:
calculating the distance between a point and the pixel coordinates of all other points by utilizing a hamming distance formula, and sequencing all the distances from small to large;
selecting the first K points with the smallest distance, inputting the first K points into a trained KNN model, respectively calculating the similarity between the first K points and the K points, and rejecting the points with the similarity higher than a preset threshold value as repeated points;
performing de-duplication treatment on repeated points in all points by using a trained KNN model and a hamming distance formula;
the pixel coordinates of all the points after the duplication removal are stored and output, so that the accurate positioning of all the elements is completed;
the method aims at the large-format PCB panel with repeated and large number of tiny components, and the defects of inaccurate positioning, poor precision and poor adaptability of the traditional machine vision positioning algorithm can be effectively avoided; according to the method, coarse positioning of Mark points is carried out by pyramid downsampling and Hough circle detection, fine positioning of sub-pixel levels is carried out by using a calliper circle detection algorithm, coordinates of a point set under a DXF coordinate system are converted into pulse coordinates, photographing position information is combined and converted into pixel coordinates to obtain pixel coordinates of a component coordinate set, repeated component coordinate information adjacent to the left side and the right side of the FOV is subjected to deduplication through a KNN algorithm, and therefore high-accuracy, high-precision and strong-adaptability rapid positioning detection is achieved, meanwhile, accurate positioning and detection can be carried out on each component on different PCB products rapidly and accurately, and the defect detection efficiency of electronic components in a PCB panel is greatly improved.
The same or similar reference numerals correspond to the same or similar components;
the terms describing the positional relationship in the drawings are merely illustrative, and are not to be construed as limiting the present patent;
it is to be understood that the above examples of the present invention are provided by way of illustration only and not by way of limitation of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.
Claims (10)
1. The large-format PCB panel accurate positioning method with the repeated elements is characterized by comprising the following steps of:
s1: acquiring initial pixel coordinates and initial pulse coordinates of Mark points on a large-format PCB panel with repeated elements by using an industrial camera and a calibration plate through a nine-point calibration method, and acquiring a homography matrix between a pulse coordinate system and a pixel coordinate system, and initial slopes and initial angles between two Mark points at the upper left and the lower right;
s2: obtaining DXF image data of a PCB panel, setting a plurality of photographing bits in the DXF image data, inputting a homography matrix into an industrial camera and initializing all parameters of the homography matrix;
s3: preprocessing the PCB image data obtained by shooting at each shooting position, respectively performing coarse positioning and fine positioning on two Mark points at the upper left and lower right in the preprocessed PCB image data, and obtaining pixel coordinates of the two Mark points at the upper left and lower right under a pixel coordinate system;
s4: converting pixel coordinates of the upper left Mark point and the lower right Mark point under a pixel coordinate system into pulse coordinates by utilizing a homography matrix, calculating the slope and the angle between the upper left Mark point and the lower right Mark point again, and rotating the preprocessed DXF image data by taking the center of the DXF coordinate system as a rotation center, so that the slope between the upper left Mark point and the lower right Mark point is the same as the initial slope, and completing the alignment of the coordinate system;
s5: according to a preset photographing position, using the initialized industrial camera to perform point set photographing, setting a plurality of fields of view in each photographing position, outputting pulse coordinates of points in the field of view after one field of view is photographed, and removing the points in the field of view from a total point set in DXF image data;
after traversing the total point set in the DXF image data, acquiring pulse coordinates of all points, and converting the pulse coordinates of each point into pixel coordinates by utilizing a homography matrix;
s6: and (3) performing de-duplication on repeated points in the traversal process of the DXF image data total point set by using a KNN algorithm, and storing and outputting pixel coordinates of all the de-duplicated points to finish accurate positioning of all elements.
2. The method for precisely positioning the large-format PCB panel with the repeating element according to claim 1, wherein in the step S1, the industrial camera and the calibration board are used to obtain the initial pixel coordinates and the initial pulse coordinates of Mark points on the large-format PCB panel with the repeating element in the pulse coordinate system by using a nine-point calibration method, and the homography matrix between the initial pulse coordinates and the initial pixel coordinates is obtained, which specifically comprises the following steps:
fixing an industrial camera on AOI industrial detection equipment, wherein the height of the industrial camera is the same as that of a large-format PCB panel with repeated elements; the calibration plate is arranged below the camera, and the position area is consistent with the working area of the industrial camera scanning PCB panel of the machine;
setting the focal length of an industrial camera, photographing, carrying out gray scale processing on the photographed image, identifying initial pixel coordinates of all Mark points by adopting a contour rounding method, generating initial pulse coordinates of all Mark points on a calibration plate according to the distance between Mark points input into the calibration plate and the position information of the first Mark point, calculating and obtaining a homography matrix according to the obtained initial pixel coordinates of the Mark points and the initial pulse coordinates corresponding to the Mark points, and recording the pulse coordinates of the current photographing position.
3. The method for precisely positioning a large-format PCB panel with repeating elements according to claim 2, wherein the homography matrix comprises a rotation matrix and a translation matrix, specifically:
wherein, (x) c ,y c ) For the initial pixel coordinates, (x) t ,y t ) For the initial pulse coordinates, R is a rotation matrix, which satisfiesM is a translation matrix, satisfying->
According to the obtained initial pixel coordinates of Mark points and the corresponding initial pulse coordinates, calculating a homography matrix by using the following formula:
wherein, (x) c1 ,y c1 ),(x c2 ,y c2 ),(x c3 ,y c3 ) Initial pixel coordinates of the first, second and third Mark, respectively; (x) t1 ,y t1 ),(x t2 ,y t2 ),(x t3 ,y t3 ) Initial pulse coordinates of the first, second and third Mark, respectively;
and equally dividing the initial pixel coordinates and the initial pulse coordinates of the nine Mark points into three groups, calculating three times according to the formula, and taking the average value as a final homography matrix.
4. The method for precisely positioning a large-format PCB panel with repeating elements according to claim 1, wherein in step S2, the specific method for inputting the homography matrix into the industrial camera and initializing all the parameters thereof is as follows:
the homography matrix is entered into the industrial camera and initializes the focal length, principal distance, distortion coefficient, adjacent pixel distance of the industrial camera, and the actual range of the single field of view in the DXF image data.
5. The method for precisely positioning the large-format PCB panel with the repeating element according to claim 1, wherein the specific method for preprocessing the PCB image data obtained by shooting at each shooting position in step S3 is as follows:
and carrying out graying treatment on the PCB image data obtained by shooting at each shooting position, and carrying out Gaussian filtering treatment and median filtering treatment on the PCB image data subjected to the graying treatment to complete pretreatment.
6. The method for precisely positioning a large-format PCB panel with repeating elements according to claim 5, wherein in step S3, the specific method for roughly positioning two Mark points on the upper left and lower right in the preprocessed PCB image data is as follows:
performing pyramid downsampling processing for 3 times on surrounding areas of upper left Mark points and lower right Mark points in the preprocessed PCB image data;
according to Hough circle detection, the first stage is used for detecting circle centers of upper left Mark points and lower right Mark points, the second stage derives circle radiuses from the circle centers, and coarse positioning of the upper left Mark points and the lower right Mark points is completed through the two stages.
7. The method for precisely positioning the large-format PCB panel with the repeating element according to claim 1 or 6, wherein in the step S3, the specific method for precisely positioning the two Mark points on the upper left and the lower right in the preprocessed PCB image data is as follows:
and (3) carrying out Gaussian filtering on the outlines of the two Mark points at the upper left and the lower right after coarse positioning by using a caliper rule detection circle method, extracting edge points, finishing fine positioning, and obtaining pixel coordinates of the two Mark points at the upper left and the lower right under a pixel coordinate system.
8. The method for precisely positioning a large-format PCB panel with repeating elements of claim 7, wherein the precision of the precise positioning is sub-pixel level.
9. The method for precisely positioning the large-format PCB panel with the repeating element according to claim 8, wherein in the step S4, the preprocessed DXF image data is rotated with the DXF coordinate system center as the rotation center, specifically:
the initial slope and the initial angle of the upper left Mark point and the lower right Mark point are respectively recorded as k 0 And a 0 The calculated slope and angle are respectively recorded as k 1 And a 1 ；
Rotating the preprocessed DXF image data by taking the center of a DXF coordinate system as a rotation center, wherein the rotation angle is a 1 -a 0 Such that the slope k between the upper left and lower right Mark points and the initial slope k 0 The same is done for the alignment of the coordinate system.
10. The method for precisely positioning the large-format PCB panel with the repeating element according to claim 9, wherein in the step S6, the specific method for performing de-duplication on the repeating points in the traversal process of the DXF image data total point set by using KNN algorithm is as follows:
calculating the distance between a point and the pixel coordinates of all other points by utilizing a hamming distance formula, and sequencing all the distances from small to large;
selecting the first K points with the smallest distance, inputting the first K points into a trained KNN model, respectively calculating the similarity between the first K points and the K points, and rejecting the points with the similarity higher than a preset threshold value as repeated points;
and performing de-duplication treatment on repeated points in all points by using the trained KNN model and a hamming distance formula.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310418227.7A CN116612071A (en) | 2023-04-18 | 2023-04-18 | Accurate positioning method for large-format PCB panel with repeated elements |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310418227.7A CN116612071A (en) | 2023-04-18 | 2023-04-18 | Accurate positioning method for large-format PCB panel with repeated elements |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116612071A true CN116612071A (en) | 2023-08-18 |
Family
ID=87678948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310418227.7A Pending CN116612071A (en) | 2023-04-18 | 2023-04-18 | Accurate positioning method for large-format PCB panel with repeated elements |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116612071A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117190866A (en) * | 2023-11-08 | 2023-12-08 | 广东工业大学 | Polarity discrimination detection method, device and equipment for multiple stacked electronic components |
-
2023
- 2023-04-18 CN CN202310418227.7A patent/CN116612071A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117190866A (en) * | 2023-11-08 | 2023-12-08 | 广东工业大学 | Polarity discrimination detection method, device and equipment for multiple stacked electronic components |
| CN117190866B (en) * | 2023-11-08 | 2024-01-26 | 广东工业大学 | Polarity discrimination detection method, device and equipment for multiple stacked electronic components |
| US12039747B1 (en) | 2023-11-08 | 2024-07-16 | Guangdong University Of Technology | Polarity discrimination detection method and apparatus for multiple stacked electronic components and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110068270B (en) | Monocular vision box volume measuring method based on multi-line structured light image recognition | |
| Bose et al. | Design of fiducials for accurate registration using machine vision | |
| CN105261022B (en) | PCB board matching method and device based on outer contour | |
| CN109211198B (en) | Intelligent target detection and measurement system and method based on trinocular vision | |
| WO2007015059A1 (en) | Method and system for three-dimensional data capture | |
| US4924506A (en) | Method for directly measuring area and volume using binocular stereo vision | |
| US5185811A (en) | Automated visual inspection of electronic component leads prior to placement | |
| EP0157299B1 (en) | Image processing apparatus | |
| CN113324478A (en) | Center extraction method of line structured light and three-dimensional measurement method of forge piece | |
| CN110827357A (en) | Combined pattern calibration board and structured light camera parameter calibration method | |
| KR20120054689A (en) | Inspection method | |
| Guerra et al. | A three-dimensional automated visual inspection system for SMT assembly | |
| CN116612071A (en) | Accurate positioning method for large-format PCB panel with repeated elements | |
| CN106952262B (en) | Ship plate machining precision analysis method based on stereoscopic vision | |
| CN112161997A (en) | Online precise visual measurement method and system for three-dimensional geometric dimension of semiconductor chip pin | |
| CN113589263B (en) | Method and system for jointly calibrating multiple homologous sensors | |
| CN114022439B (en) | Flexible circuit board defect detection method based on morphological image processing | |
| CN115456945A (en) | Chip pin defect detection method, detection device and equipment | |
| KR20240040708A (en) | Board measurement system and method thereof | |
| CN110533774A (en) | A kind of method for reconstructing three-dimensional model based on smart phone | |
| WO2019099544A1 (en) | Geometric calibration of x-ray imaging systems | |
| CN111368573A (en) | Positioning method based on geometric feature constraint | |
| US6898333B1 (en) | Methods and apparatus for determining the orientation of an object in an image | |
| CN113112396A (en) | Method for detecting conductive particles | |
| CN111260735B (en) | External parameter calibration method for single-shot LIDAR and panoramic camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |