CN116008177A - SMT component high defect identification method, system and readable medium thereof - Google Patents
SMT component high defect identification method, system and readable medium thereof Download PDFInfo
- Publication number
- CN116008177A CN116008177A CN202211561975.2A CN202211561975A CN116008177A CN 116008177 A CN116008177 A CN 116008177A CN 202211561975 A CN202211561975 A CN 202211561975A CN 116008177 A CN116008177 A CN 116008177A
- Authority
- CN
- China
- Prior art keywords
- height
- smt
- smt component
- component
- circuit board
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000009466 transformation Effects 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 12
- 230000010363 phase shift Effects 0.000 claims description 8
- 238000013519 translation Methods 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000003491 array Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 16
- 238000005516 engineering process Methods 0.000 abstract description 15
- 238000010586 diagram Methods 0.000 description 3
- 238000005476 soldering Methods 0.000 description 3
- 238000003466 welding Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical compound [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
Images
Classifications
-
- 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
- Image Processing (AREA)
Abstract
The invention discloses a method, a system and a readable medium for identifying the height defect of an SMT (surface mounted technology) component, which are characterized in that a three-dimensional point cloud model image of the SMT circuit board to be tested is obtained, and the positioning information of the SMT component on the SMT circuit board is matched and inquired through an ICP algorithm; setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image; and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not. The method and the device can be used for identifying and detecting the defects of the SMT components in the height direction such as the height direction and the tilting direction, and can be used for realizing comprehensive and high-precision automatic defect detection of the SMT patch by combining the existing two-dimensional surface detection technology.
Description
Technical Field
The invention relates to the technical field of STM patches, in particular to a method and a system for identifying high defects of SMT components and a readable medium thereof.
Background
SMT is a surface mount technology (surface mount technology) (Surface Mounted Technology abbreviation) and is currently one of the most popular technologies and techniques in the electronics assembly industry. Electronic circuit surface mount technology (Surface Mount Technology, SMT) is known as surface mount or surface mount technology. The circuit mounting technology is to mount no-pin or short-lead surface assembly components (SMC/SMD, chinese called chip components) on the surface of a circuit board (Printed Circuit Board, PCB) or the surface of other substrates on which the integrated STM components are printed, and to perform welding assembly by reflow soldering or dip soldering and other methods.
The SMT technology replaces manual sticking, however, some defective products appear in the welding process of a chip mounter, so that the detection technology is extremely important, an efficient detection system can greatly reduce the factory return rate, effectively improve the industrial production efficiency, common SMT sticking defects are divided into two types, and the first type of defects comprise defects in 2D ranges such as component offset, standing pieces, foreign matters, component sticking inverse and the like; the second type of defect includes defects in the 3D range of component thickness errors, trace bend offsets, trace intersections, and the like. In order to detect the first kind of defects, the AOI system needs to splice the obtained high-resolution SMT patch local images in real time, and uses machine vision theory knowledge to perform positioning detection on the spliced panoramic image. The AOI principle is that three colors of RGB light are irradiated onto an element from three different angles and then reflected back, different welding spot forms reflect different color lights, and vector analysis is carried out on data such as color, brightness and the like of the reflected light by reference positions and set parameters. The detection capability has a certain limitation, only two-dimensional image information can be adopted, the height of the components can not be obtained,
the existing 2D detection means is difficult to identify the height defects of SMT components, the defects can generate serious faults such as short circuit, cold joint and the like, the three-dimensional detection can accurately detect the heights of the components and soldering tin, further the components are further analyzed, and the detection accuracy can be greatly improved; therefore, there is a need to develop a method for identifying high defects of SMT components.
Disclosure of Invention
The invention aims to solve the technical problem that the conventional 2D detection means are difficult to identify the height defect of an SMT component.
In order to solve the above technical problems, a first aspect of the present invention provides a method for identifying a height defect of an SMT component, the method comprising:
acquiring a three-dimensional point cloud model image of an SMT circuit board to be tested, and searching positioning information of an SMT component on the SMT circuit board in a matching way through an ICP algorithm;
setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image;
and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not.
Further, the method further comprises:
acquiring a three-dimensional point cloud model image of a qualified SMT circuit board in advance, and matching and inquiring positioning information of an SMT component on the qualified SMT circuit board through an ICP algorithm;
and setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring and storing the height threshold value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image.
Further, the obtaining the three-dimensional point cloud model image of the SMT circuit board to be tested includes:
purifying SMT circuit board pictures to be tested with different visual angles through a RANSAC algorithm to obtain feature matching points, and setting feature point sets of two point clouds as So and X respectively;
setting the iteration times as k, enabling k=0, and performing initial transformation on the purified characteristic point set So by using a space transformation matrix Ro and To calculated by the RANSAC To establish a Kd-tree of the characteristic point set X;
S 1 ＝R 0 S 0 +T 0 ，
finding the nearest point Sk1 of Sk in X, calculating coordinate transformation know arrays Rk and Tk by using feature matching point sets Sk and Sk1, and carrying out coordinate transformation of the feature point sets by using the following formula:
S k+1 ＝R k S k +T k
it is determined whether the distance error D converges, if Dk-dk+1<M, then it converges, where M is a set threshold and M >0, otherwise, it re-finds the nearest point Sk1 of Sk in X using the feature matching point sets Sk and Sk1.
Further, the purifying the SMT circuit board patterns to be tested with different view angles through the RANSAC algorithm to obtain the feature matching points comprises the following steps:
performing feature matching on at least two SMT circuit board pictures to be tested with different visual angles by using a SIFT algorithm, setting a threshold value 0.6 for judging matching, and finding N1 pairs of feature matching points;
and simultaneously setting an identification area of A, finding N2 pairs of feature matching points altogether, calculating the N2 pairs of feature matching points by using a RANSAC algorithm, and setting a threshold value To be 0.8, namely at least 80% of feature points are required To meet the solved rotation matrix and translation vector, and obtaining the rotation matrix Ro and translation vector To.
Further, setting a plurality of identification areas on the SMT component according to the positioning information, and obtaining a height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image specifically includes:
and setting at least 4 identification areas on the upper, lower, left and right sides of the SMT component plane according to the positioning information, extracting the height values of all feature matching points in each identification area according to the relation of the SMT component areas in the three-dimensional point cloud model image, and taking the average value of the height values of all feature matching points as the height value of the identification area.
Further, comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect comprises:
setting a minimum height threshold value Hl and a maximum height threshold value Hh of the SMT component, comparing the height value of the identification area with the minimum height threshold value Hl and the maximum height threshold value Hh, and judging that the element height is low if the height value is lower than the minimum height threshold value Hl and judging that the element height is high if the height value is higher than the maximum height threshold value Hh.
Further, the comparing the height value of the identification area with a preset height threshold value, and determining whether the SMT component has a height defect further includes:
setting a horizontal inclination angle threshold TH, acquiring a horizontal inclination angle T1 according to the height difference between the identification areas in the horizontal direction of the SMT component, and judging that the component is inclined horizontally when the horizontal inclination angle T1 of the SMT component is larger than the horizontal inclination angle threshold TH;
setting a vertical inclination angle threshold value TV, acquiring a vertical inclination angle T2 according to the height difference between the identification areas in the vertical direction of the SMT component, and judging that the component is vertical inclined when the vertical inclination angle T2 of the SMT component is larger than the vertical inclination angle threshold value TV.
The second aspect of the present invention provides a SMT component height defect recognition system, comprising:
the system comprises a group of CCD cameras and projectors, wherein the CCD cameras and the projectors are arranged on a workbench surface, the CCD cameras are used for shooting patterns of light projected to the surface of an SMT circuit board to be tested by the projectors, and three-dimensional point cloud model images of the SMT circuit board to be tested are obtained through system calibration parameters and phase-height mapping relations;
the graphics processing terminal is used for receiving and storing the three-dimensional point cloud model image of the SMT circuit board to be tested, and searching the positioning information of the SMT component on the SMT circuit board in a matching way through an ICP algorithm; setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image; and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not.
Furthermore, calibrating the system according to the SMT structured light combined calibration model, obtaining a relative phase through n-step phase shift calculation of the graph on the surface of the SMT circuit board to be tested, unfolding the relative phase by utilizing a multi-frequency heterodyne phase unwrapping method to obtain an absolute phase, and obtaining a three-dimensional point cloud model image of the SMT circuit board to be tested according to the correspondence between the absolute phase and the surface height of the SMT circuit board to be tested.
A third aspect of the present invention provides a computer readable storage medium for storing program data which, when executed by a processor, is configured to implement an SMT component height defect recognition method as described above.
The technical scheme of the invention has the beneficial effects that:
according to the SMT component height defect identification method, the SMT component height defect identification system and the SMT component height defect identification readable medium, the three-dimensional point cloud model image of the SMT circuit board to be detected is obtained, and the positioning information of the SMT component on the SMT circuit board is matched and inquired through an ICP algorithm; setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image; and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not. The method and the device can be used for identifying and detecting the defects of the SMT components in the height direction such as the height direction and the tilting direction, and can be used for realizing comprehensive and high-precision automatic defect detection of the SMT patch by combining the existing two-dimensional surface detection technology.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for identifying high defects of SMT components according to an embodiment of the invention;
FIG. 2 is a flow chart of a method for establishing a height threshold according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a hardware architecture of an SMT component height defect recognition system according to an embodiment of the invention;
FIG. 4 is a schematic diagram of a binocular camera calibration system model of an embodiment of the present invention;
FIG. 5 is a schematic diagram of a multi-frequency heterodyne method in accordance with an embodiment of the present invention;
FIG. 6 is a flow chart of a three-frequency heterodyne phase unwrapping process in accordance with an embodiment of the present invention;
Detailed Description
The following describes the embodiments of the present invention further with reference to the drawings. The description of these embodiments is provided to assist understanding of the present invention, but is not intended to limit the present invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
Example 1
As shown in fig. 1, an embodiment of the present invention provides a method for identifying a high defect of an SMT component, the method comprising:
s101, acquiring a three-dimensional point cloud model image of an SMT circuit board to be tested, and matching and inquiring positioning information of an SMT component on the SMT circuit board through an ICP algorithm;
specifically, the obtaining the three-dimensional point cloud model image of the SMT circuit board to be tested includes:
purifying SMT circuit board pictures to be tested with different visual angles through a RANSAC algorithm to obtain feature matching points, and setting feature point sets of two point clouds as So and X respectively;
setting the iteration times as k, enabling k=0, and performing initial transformation on the purified characteristic point set So by using a space transformation matrix Ro and To calculated by the RANSAC To establish a Kd-tree of the characteristic point set X;
S 1 ＝R 0 S 0 +T 0 ，
finding the nearest point Sk1 of Sk in X, calculating coordinate transformation know arrays Rk and Tk by using feature matching point sets Sk and Sk1, and carrying out coordinate transformation of the feature point sets by using the following formula:
S k+1 ＝R k S k +T k
it is determined whether the distance error D converges, if Dk-dk+1<M, then it converges, where M is a set threshold and M >0, otherwise, it re-finds the nearest point Sk1 of Sk in X using the feature matching point sets Sk and Sk1.
Feature matching points purified by the RANSAC technology are achieved, and statistical solution is carried out on the ICP algorithm. Because the initial values of the rotation matrix and the translation vector established before have great accuracy, the ICP algorithm only obtains a more accurate rotation matrix R and the translation vector T through optimizing the initial values. The point clouds obtained from different visual angles can be matched, and three-dimensional reconstruction of the three-dimensional SMT patch and matching of three-dimensional components are realized.
Specifically, the purifying the SMT circuit board patterns to be tested with different visual angles through the RANSAC algorithm to obtain the feature matching points comprises the following steps:
performing feature matching on at least two SMT circuit board pictures to be tested with different visual angles by using a SIFT algorithm, setting a threshold value 0.6 for judging matching, and finding N1 pairs of feature matching points;
and simultaneously setting an identification area of A, finding N2 pairs of feature matching points altogether, calculating the N2 pairs of feature matching points by using a RANSAC algorithm, and setting a threshold value To be 0.8, namely at least 80% of feature points are required To meet the solved rotation matrix and translation vector, and obtaining the rotation matrix Ro and translation vector To.
S102, setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring a height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image;
specifically, the setting a plurality of identification areas on the SMT component according to the positioning information, and obtaining a height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image specifically includes:
and setting at least 4 identification areas on the upper, lower, left and right sides of the SMT component plane according to the positioning information, extracting the height values of all feature matching points in each identification area according to the relation of the SMT component areas in the three-dimensional point cloud model image, and taking the average value of the height values of all feature matching points as the height value of the identification area.
S103, comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not.
Specifically, the comparing, according to the height value of the identification area with a preset height threshold, whether the SMT component has a height defect includes:
setting a minimum height threshold value Hl and a maximum height threshold value Hh of the SMT component, comparing the height value of the identification area with the minimum height threshold value Hl and the maximum height threshold value Hh, and judging that the element height is low if the height value is lower than the minimum height threshold value Hl and judging that the element height is high if the height value is higher than the maximum height threshold value Hh.
Specifically, the comparing, according to the height value of the identification area and the preset height threshold, the determining whether the SMT component has the height defect further includes:
setting a horizontal inclination angle threshold TH, acquiring a horizontal inclination angle T1 according to the height difference between the identification areas in the horizontal direction of the SMT component, and judging that the component is inclined horizontally when the horizontal inclination angle T1 of the SMT component is larger than the horizontal inclination angle threshold TH;
setting a vertical inclination angle threshold value TV, acquiring a vertical inclination angle T2 according to the height difference between the identification areas in the vertical direction of the SMT component, and judging that the component is vertical inclined when the vertical inclination angle T2 of the SMT component is larger than the vertical inclination angle threshold value TV.
As shown in fig. 2, the method further includes:
s201, acquiring a three-dimensional point cloud model image of a qualified SMT circuit board in advance, and matching and inquiring positioning information of an SMT component on the qualified SMT circuit board through an ICP algorithm;
s202, setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring and storing the height threshold value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image.
Example 2
As shown in fig. 3, the embodiment of the invention further provides a SMT component height defect recognition system, which includes:
the system comprises a group of CCD cameras 10 and a projector 20 arranged on a workbench surface 40, wherein the CCD cameras 10 are used for shooting a graph of light projected by the projector 20 onto the surface of an SMT circuit board 50 to be tested, and a three-dimensional point cloud model image of the SMT circuit board 50 to be tested is obtained through system calibration parameters and a phase-height mapping relation;
the graphic processing terminal 30 receives and stores the three-dimensional point cloud model image of the SMT circuit board 50 to be tested, and matches and inquires the positioning information of the SMT components on the SMT circuit board through an ICP algorithm; setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image; and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not.
And calibrating the system according to the SMT structured light combined calibration model, obtaining a relative phase by n-step phase shift calculation of the graph on the surface of the SMT circuit board to be tested, obtaining an absolute phase by using a multi-frequency heterodyne phase unwrapping method to unwrap the relative phase, and obtaining a three-dimensional point cloud model image of the SMT circuit board to be tested according to the absolute phase and the correspondence of the surface height of the SMT circuit board to be tested.
Specifically, a binocular camera is adopted to collect three-dimensional information of an object at the same time, and the mapping relation of two image coordinate systems is solved by using parameters and other constraint conditions between the two images. However, in the integrated system, the two coordinate systems belong to different coordinate systems, and according to the correspondence between the absolute phase and the surface height of the measured object, the three-dimensional image of the measured object can be obtained from the absolute phase.
But in the integrated system the two belong to different coordinate systems, as shown in FIG. 4, Q L -X L Y L Z L For the left camera coordinate system, Q C -X C Y C Z C In the coordinate system of the right camera, the coordinate system of the left camera is the coordinate system of the right camera,the conversion between the pixel point coordinates (u, v) of the left camera image plane and the coordinates (xl, yl, zl) of the marker points in the right camera coordinate system is as follows:
the method comprises the steps of obtaining three-dimensional object points of each combined photographed picture and two-dimensional projection coordinates thereof, namely world coordinates and pixel coordinates corresponding to chessboard angular points, establishing a spatial transformation relation by the formulas, and calculating a coordinate transformation matrix between the two cameras.
To obtain the surface shape data of a three-dimensional object, a surface-structured light must be projected onto the surface of the object to be measured. In the phase detection profile technique, a light intensity distribution is generally regarded as a grating image with standard sinusoidal distribution, which is regarded as a planar structure light path. When the grating is projected on the three-dimensional object surface, the projected object is modulated by the sinusoidal grating, and the light intensity is expressed as follows:
where Ii (x, y) is the light intensity corresponding to the ith interference fringe, δi is the ith phase shift, a (x, y) is the background light intensity, B (x, y) is the modulation depth of the fringe, Φ (x, y) is the phase to be measured. A (x, y), B (x, y) and Φ (x, y) are three unknowns, so at least three images are required to calculate Φ (x, y).
The relative displacements chosen for the three raster images are assumed to be: -2pi/3,0,2 pi/3, then its relative light intensity expressions are respectively:
wherein A represents the background value of the natural light intensity of the image, B represents the background value of the light intensity of the projected stripes, and the above formula is developed as follows:
the further conversion is as follows:
the above is the case where the phase shift is selected to be-2 pi/3,0,2 pi/3, and the selection of an appropriate phase shift can simplify the calculation. The three-step phase shift method has small number of images to be collected and high data processing speed.
When we calculate the three-dimensional height information of the points, we need the absolute phase of the object height, but we can not obtain the absolute phase of the object by the phase shift method, but can only obtain the corresponding relative phase, which is a discontinuous function with continuous 2pi as the period. Therefore, after the relative phase is obtained, three-dimensional information of a three-dimensional point in space cannot be directly calculated, and the relative phase value needs to be processed to obtain an absolute phase value corresponding to the height information one by one, which is called phase unwrapping or phase unwrapping.
In the embodiment of the application, the experimental calculation of phase unwrapping is performed by using a multi-frequency heterodyne phase unwrapping method, and the basic principle of multi-frequency heterodyne is that different frequency phase functions Φ1 (X) and Φ2 are superimposed into a lower frequency phase function Φ (X), as shown in fig. 5, where λ1, λ2, and λb are the frequencies of the phase functions Φ1 (X), Φ2, and Φ (X), respectively, and λb can be calculated from λ1 and λ2 to obtain a new frequency λb:
new phase Φb:
the multi-frequency heterodyne method adopts the integration and difference criteria of trigonometric functions, and the total detection range is in the same period of a lower frequency spectrum signal because the frequency spectrum obtained by superposition is lower, so that the absolute interference phase is determined.
The measurement is carried out according to the gratings with three frequencies, and the gratings phi 1, phi 2 and phi 3 with three different frequencies are used; as shown in fig. 6, Φ12 and Φ23 are obtained from Φ1- Φ2 and Φ2- Φ3, and finally, a grating with a frequency of 1 is generated, and finally, continuous absolute phases are obtained, so that calculation from relative phase to absolute phase is realized.
Example 3
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with an SMT component height defect recognition program, and the SMT component height defect recognition program realizes the steps of the SMT component height defect recognition method when being executed by a processor.
It will be appreciated that the computer readable storage medium in this embodiment may be applied to a server, and specific implementation thereof may refer to the above embodiment, which is not described herein.
According to the SMT component height defect identification method, the SMT component height defect identification system and the SMT component height defect identification readable medium, the three-dimensional point cloud model image of the SMT circuit board to be detected is obtained, and the positioning information of the SMT component on the SMT circuit board is matched and inquired through an ICP algorithm; setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image; and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not. The method and the device can be used for identifying and detecting the defects of the SMT components in the height direction such as the height direction and the tilting direction, and can be used for realizing comprehensive and high-precision automatic defect detection of the SMT patch by combining the existing two-dimensional surface detection technology.
Claims (10)
1. A method for identifying a high defect of an SMT component, said method comprising:
acquiring a three-dimensional point cloud model image of an SMT circuit board to be tested, and searching positioning information of an SMT component on the SMT circuit board in a matching way through an ICP algorithm;
setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image;
and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not.
2. A SMT component height defect identification method according to claim 1, wherein said method further comprises:
acquiring a three-dimensional point cloud model image of a qualified SMT circuit board in advance, and matching and inquiring positioning information of an SMT component on the qualified SMT circuit board through an ICP algorithm;
and setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring and storing the height threshold value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image.
3. The SMT component height defect identification method of claim 1, wherein said obtaining a three-dimensional point cloud model image of the SMT circuit board to be tested comprises:
purifying SMT circuit board pictures to be tested with different visual angles through a RANSAC algorithm to obtain feature matching points, and setting feature point sets of two point clouds as So and X respectively;
setting the iteration times as k, enabling k=0, and performing initial transformation on the purified characteristic point set So by using a space transformation matrix Ro and To calculated by the RANSAC To establish a Kd-tree of the characteristic point set X;
S 1 ＝R 0 S 0 +T 0 ，
finding the nearest point Sk1 of Sk in X, calculating coordinate transformation know arrays Rk and Tk by using feature matching point sets Sk and Sk1, and carrying out coordinate transformation of the feature point sets by using the following formula:
S k+1 ＝R k S k +T k
it is determined whether the distance error D converges, if Dk-dk+1<M, then it converges, where M is a set threshold and M >0, otherwise, it re-finds the nearest point Sk1 of Sk in X using the feature matching point sets Sk and Sk1.
4. The SMT component height defect identification method of claim 3, wherein said purifying SMT circuit board patterns to be tested for different viewing angles by RANSAC algorithm to obtain feature matching points comprises:
performing feature matching on at least two SMT circuit board pictures to be tested with different visual angles by using a SIFT algorithm, setting a threshold value 0.6 for judging matching, and finding N1 pairs of feature matching points;
and simultaneously setting an identification area of A, finding N2 pairs of feature matching points altogether, calculating the N2 pairs of feature matching points by using a RANSAC algorithm, and setting a threshold value To be 0.8, namely at least 80% of feature points are required To meet the solved rotation matrix and translation vector, and obtaining the rotation matrix Ro and translation vector To.
5. The SMT component height defect identification method according to claim 1, wherein said setting a plurality of identification areas on said SMT component according to said positioning information, and obtaining a height value of each of said identification areas according to said SMT component area relation in said three-dimensional point cloud model image specifically comprises:
and setting at least 4 identification areas on the upper, lower, left and right sides of the SMT component plane according to the positioning information, extracting the height values of all feature matching points in each identification area according to the relation of the SMT component areas in the three-dimensional point cloud model image, and taking the average value of the height values of all feature matching points as the height value of the identification area.
6. The SMT component height defect identification method of claim 1, wherein said determining whether said SMT component has a height defect based on a height value of said identification area compared to a predetermined height threshold value comprises:
setting a minimum height threshold value Hl and a maximum height threshold value Hh of the SMT component, comparing the height value of the identification area with the minimum height threshold value Hl and the maximum height threshold value Hh, and judging that the element height is low if the height value is lower than the minimum height threshold value Hl and judging that the element height is high if the height value is higher than the maximum height threshold value Hh.
7. The SMT component height defect identification method of claim 1, wherein said determining whether said SMT component has a height defect according to a height value of said identification area compared to a preset height threshold value further comprises:
setting a horizontal inclination angle threshold TH, acquiring a horizontal inclination angle T1 according to the height difference between the identification areas in the horizontal direction of the SMT component, and judging that the component is inclined horizontally when the horizontal inclination angle T1 of the SMT component is larger than the horizontal inclination angle threshold TH;
setting a vertical inclination angle threshold value TV, acquiring a vertical inclination angle T2 according to the height difference between the identification areas in the vertical direction of the SMT component, and judging that the component is vertical inclined when the vertical inclination angle T2 of the SMT component is larger than the vertical inclination angle threshold value TV.
8. An SMT component height defect recognition system, said system comprising:
the system comprises a group of CCD cameras and projectors, wherein the CCD cameras and the projectors are arranged on a workbench surface, the CCD cameras are used for shooting patterns of light projected to the surface of an SMT circuit board to be tested by the projectors, and three-dimensional point cloud model images of the SMT circuit board to be tested are obtained through system calibration parameters and phase-height mapping relations;
the graphics processing terminal is used for receiving and storing the three-dimensional point cloud model image of the SMT circuit board to be tested, and searching the positioning information of the SMT component on the SMT circuit board in a matching way through an ICP algorithm; setting a plurality of identification areas on the SMT component according to the positioning information, and acquiring the height value of each identification area according to the SMT component area relation in the three-dimensional point cloud model image; and comparing the height value of the identification area with a preset height threshold value, and judging whether the SMT component has a height defect or not.
9. The SMT component height defect identification system according to claim 8, wherein the system is calibrated according to the SMT structured light combined calibration model, the relative phase is obtained by n-step phase shift calculation of the pattern on the surface of the SMT circuit board to be tested, the absolute phase is obtained by performing relative phase unwrapping by using a multi-frequency heterodyne phase unwrapping method, and the three-dimensional point cloud model image of the SMT circuit board to be tested is obtained according to the correspondence between the absolute phase and the surface height of the SMT circuit board to be tested.
10. A computer readable storage medium for storing program data which, when executed by a processor, is adapted to carry out the steps of the SMT component height defect identification method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211561975.2A CN116008177A (en) | 2022-12-07 | 2022-12-07 | SMT component high defect identification method, system and readable medium thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211561975.2A CN116008177A (en) | 2022-12-07 | 2022-12-07 | SMT component high defect identification method, system and readable medium thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116008177A true CN116008177A (en) | 2023-04-25 |
Family
ID=86023834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211561975.2A Pending CN116008177A (en) | 2022-12-07 | 2022-12-07 | SMT component high defect identification method, system and readable medium thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116008177A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117670887A (en) * | 2024-02-01 | 2024-03-08 | 湘潭大学 | Tin soldering height and defect detection method based on machine vision |
-
2022
- 2022-12-07 CN CN202211561975.2A patent/CN116008177A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117670887A (en) * | 2024-02-01 | 2024-03-08 | 湘潭大学 | Tin soldering height and defect detection method based on machine vision |
| CN117670887B (en) * | 2024-02-01 | 2024-04-09 | 湘潭大学 | Tin soldering height and defect detection method based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6457072B2 (en) | Integration of point clouds from multiple cameras and light sources in 3D surface shape measurement | |
| CN111815716B (en) | Parameter calibration method and related device | |
| CN107270810B (en) | The projector calibrating method and device of multi-faceted projection | |
| Huang et al. | A fast and flexible projector-camera calibration system | |
| CN111383279A (en) | External parameter calibration method and device and electronic equipment | |
| US20240303772A1 (en) | Device and method for correspondence analysis in images | |
| CN107271445B (en) | Defect detection method and device | |
| CN115082538A (en) | System and method for three-dimensional reconstruction of surface of multi-view vision balance ring part based on line structure light projection | |
| JP6186431B2 (en) | Calibration apparatus, calibration system, and imaging apparatus | |
| CN113781434A (en) | Defect detection method and device, intelligent terminal and computer readable storage medium | |
| CN116008177A (en) | SMT component high defect identification method, system and readable medium thereof | |
| CN112361991A (en) | Three-dimensional scanning method and device, computer equipment and storage medium | |
| CN110044266B (en) | Photogrammetry system based on speckle projection | |
| CN116152066B (en) | Point cloud detection method, system, equipment and medium for complete appearance of element | |
| CN111566438B (en) | Image acquisition method and system | |
| CN114485479B (en) | Structured light scanning and measuring method and system based on binocular camera and inertial navigation | |
| CN116429014A (en) | Orthogonal multi-projection aliasing image non-separation structured light three-dimensional measurement method | |
| JP7266070B2 (en) | Board wiring measurement system and method | |
| CN114581533A (en) | Combined calibration method and device | |
| Pang et al. | Automated optical inspection of solder paste based on 2.5 D visual images | |
| CN108759720B (en) | Smooth surface profile measuring method | |
| CN112037178B (en) | Cylindrical two-dimensional image generation method based on multi-view camera | |
| Svetkoff et al. | Automatic Inspection of Component Boards Using 3‐D and Greyscale Vision | |
| Drouin1&2 et al. | An energy formulation for establishing the correspondence used in projector calibration | |
| KR101765257B1 (en) | Method for acquiring three dimensional image information, and computing device implementing the samemethod |
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 |