CN118052816A - PCBA surface defect detection method and system - Google Patents
PCBA surface defect detection method and system Download PDFInfo
- Publication number
- CN118052816A CN118052816A CN202410446300.6A CN202410446300A CN118052816A CN 118052816 A CN118052816 A CN 118052816A CN 202410446300 A CN202410446300 A CN 202410446300A CN 118052816 A CN118052816 A CN 118052816A
- Authority
- CN
- China
- Prior art keywords
- image
- difference
- target
- information
- pixel point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 76
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- RVCKCEDKBVEEHL-UHFFFAOYSA-N 2,3,4,5,6-pentachlorobenzyl alcohol Chemical compound OCC1=C(Cl)C(Cl)=C(Cl)C(Cl)=C1Cl RVCKCEDKBVEEHL-UHFFFAOYSA-N 0.000 title claims abstract 5
- 238000013507 mapping Methods 0.000 claims abstract description 98
- 238000012545 processing Methods 0.000 claims abstract description 43
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000012216 screening Methods 0.000 claims abstract description 24
- 238000010586 diagram Methods 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000003466 welding Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000005187 foaming Methods 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The application relates to a PCBA surface defect detection method and system. The method comprises the following steps: obtaining standard structure images of all angles of a standard circuit board and structure images to be detected of all angles of the circuit board to be detected, and carrying out image mapping processing on the standard structure images of all angles and the structure images to be detected of all angles according to each angle to obtain a mapping comparison graph of all angles; identifying differential pixel points in the mapping contrast diagram through an image differential identification network, and clustering each differential pixel point to obtain each differential image; and identifying the image difference type of each difference image, and screening a target difference image of the target image difference type and point position information of each difference pixel point of the target difference image as target defect information of the circuit board to be tested. By adopting the method, the accuracy of PCBA surface defect detection can be improved.
Description
Technical Field
The application relates to the technical field of artificial intelligence, in particular to a PCBA surface defect detection method and system.
Background
PCBA inspection refers to a series of tests performed on PCBA (printed circuit board) circuit boards with electronic components mounted to ensure their electrical continuity, functionality and reliability. But is particularly important for PCBA surface detection, so how to promote intelligent detection of PCBA surface defects is the current research focus.
The PCBA surface defect detection mode in the traditional technology is a scheme of detecting through shooting images and through the images, but the PCBA surface defect detection accuracy is low due to the fact that defects exist in image definition and high-precision image detection efficiency is low.
Disclosure of Invention
Accordingly, it is desirable to provide a method and a system for detecting surface defects of PCBA, in order to solve the above-mentioned problems.
In a first aspect, the application provides a PCBA surface defect detection method. The method comprises the following steps:
obtaining standard structure images of all angles of a standard circuit board and structure images to be detected of all angles of the circuit board to be detected, and carrying out image mapping processing on the standard structure images of all angles and the structure images to be detected of all angles according to each angle to obtain a mapping comparison graph of all angles;
Identifying differential pixel points in the mapping contrast diagram through an image differential identification network, and clustering each differential pixel point to obtain each differential image;
And identifying the image difference type of each difference image, and screening a target difference image of the target image difference type and point position information of each difference pixel point of the target difference image as target defect information of the circuit board to be tested.
Optionally, the performing image mapping processing on the standard structural image of the angle and the structural image to be detected of the angle to obtain a mapping contrast diagram of the angle includes:
Performing equidistant edge marking processing on each structural image to obtain each edge position point of each structural image;
And carrying out overlapping mapping processing on each edge position point based on the edge identification information of each edge position point to obtain a mapping comparison chart.
Optionally, before the identifying, by the image difference identifying network, the difference pixel point in the mapping contrast chart further includes:
Gridding the mapping comparison graph to obtain each sub-mapping image of the mapping comparison graph;
And carrying out pixel point marking processing on each pixel point in each sub-mapping image by a pixel point marking strategy according to each sub-mapping image to obtain point identification information of each pixel point.
Optionally, the identifying, by the image difference identifying network, the difference pixel point in the map contrast map includes:
inputting a first pixel point of a standard structure image with the same point identification information and a second pixel point of a structure image to be detected into an image difference identification network to obtain a difference identification result between each first pixel point and each second pixel point;
and taking the first pixel point and the second pixel point which are the difference of the difference recognition result as the difference pixel points in the mapping contrast diagram.
Optionally, the clustering processing is performed on each difference pixel point to obtain each difference image, including:
identifying pixel point distance values between the different pixel points based on the point identification information of the different pixel points and the sub-mapping image to which the different pixel points belong;
And clustering each difference pixel point and each pixel point between each difference pixel point based on the pixel point distance value between each difference pixel point to obtain each difference image.
Optionally, before identifying the image difference type of each difference image, the method further includes:
Identifying each plate structure of the standard circuit board and the position range of each plate structure in the structural image, and collecting the structural range ratio of each sub-structural information of each plate structure and the structural position information of each sub-structural information;
Identifying a sub-positional range of each piece of sub-structural information in a structural range of the plate structure based on a structural range ratio of each piece of sub-structural information and structural positional information of each piece of sub-structural information;
and identifying the image position range of each sub-map image in the structural image, and determining the sub-structure information corresponding to each sub-map image based on the sub-position range to which each image position range belongs.
Optionally, the identifying the image difference type of each difference image includes:
For each difference image, identifying each target sub-map image corresponding to the difference image, and identifying target sub-structure information corresponding to the difference image based on sub-structure information corresponding to each target sub-map image;
and inquiring the surface defect type of the target substructure information in a database, and identifying the image difference type of the difference image through an image identification network based on the surface defect type of the target substructure information.
Optionally, the screening the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image, as the target defect information of the circuit board to be tested, includes:
Responding to the data uploading operation of the staff, acquiring a difference type of the target image, and screening the difference image of the difference type of the target image from the difference images to serve as the target difference image;
identifying, for each target difference image, point location information for each difference pixel point based on an image location range of the target difference image in the structural image and point identification information for each difference pixel point in the target difference image;
And taking the image difference type of each target difference image and the point position information of each difference pixel point of each target difference image as target defect information of the circuit board to be tested.
In a second aspect, the application further provides a PCBA surface defect detection system. The system comprises:
The acquisition module is used for acquiring standard structure images of all angles of the standard circuit board and to-be-detected structure images of all angles of the circuit board to be detected, and carrying out image mapping processing on the standard structure images of the angles and the to-be-detected structure images of the angles aiming at each angle to obtain a mapping comparison graph of the angles;
The identification module is used for identifying the difference pixel points in the mapping contrast diagram through an image difference identification network, and carrying out clustering treatment on each difference pixel point to obtain each difference image;
The screening module is used for identifying the image difference type of each difference image, screening the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image, and taking the point position information as the target defect information of the circuit board to be tested.
Optionally, the acquiring module is specifically configured to:
Performing equidistant edge marking processing on each structural image to obtain each edge position point of each structural image;
And carrying out overlapping mapping processing on each edge position point based on the edge identification information of each edge position point to obtain a mapping comparison chart.
Optionally, the system further comprises:
the mapping module is used for carrying out gridding treatment on the mapping comparison graph to obtain each sub-mapping image of the mapping comparison graph;
the marking module is used for carrying out pixel marking processing on each pixel point in each sub-mapping image through a pixel point marking strategy to obtain point identification information of each pixel point.
Optionally, the identification module is specifically configured to:
inputting a first pixel point of a standard structure image with the same point identification information and a second pixel point of a structure image to be detected into an image difference identification network to obtain a difference identification result between each first pixel point and each second pixel point;
and taking the first pixel point and the second pixel point which are the difference of the difference recognition result as the difference pixel points in the mapping contrast diagram.
Optionally, the identification module is specifically configured to:
identifying pixel point distance values between the different pixel points based on the point identification information of the different pixel points and the sub-mapping image to which the different pixel points belong;
And clustering each difference pixel point and each pixel point between each difference pixel point based on the pixel point distance value between each difference pixel point to obtain each difference image.
Optionally, the system further includes:
The acquisition module is used for identifying each plate structure of the standard circuit board and the position range of each plate structure in the structural image, and acquiring the structural range ratio of each sub-structural information of each plate structure and the structural position information of each sub-structural information;
A range identifying module, configured to identify, in a structural range of the board structure, a sub-position range of each piece of sub-structure information based on a structural range ratio of each piece of sub-structure information and structural position information of each piece of sub-structure information;
The determining module is used for identifying the image position range of each sub-map image in the structural image and determining the sub-structure information corresponding to each sub-map image based on the sub-position range to which each image position range belongs.
Optionally, the screening module is specifically configured to:
For each difference image, identifying each target sub-map image corresponding to the difference image, and identifying target sub-structure information corresponding to the difference image based on sub-structure information corresponding to each target sub-map image;
and inquiring the surface defect type of the target substructure information in a database, and identifying the image difference type of the difference image through an image identification network based on the surface defect type of the target substructure information.
Optionally, the screening module is specifically configured to:
Responding to the data uploading operation of the staff, acquiring a difference type of the target image, and screening the difference image of the difference type of the target image from the difference images to serve as the target difference image;
identifying, for each target difference image, point location information for each difference pixel point based on an image location range of the target difference image in the structural image and point identification information for each difference pixel point in the target difference image;
And taking the image difference type of each target difference image and the point position information of each difference pixel point of each target difference image as target defect information of the circuit board to be tested.
According to the PCBA surface defect detection method and system, the standard structure images of all angles of the standard circuit board and the structure images to be detected of all angles of the circuit board to be detected are obtained, and for each angle, the standard structure images of the angles and the structure images to be detected of the angles are subjected to image mapping processing to obtain the mapping comparison graph of the angles; identifying differential pixel points in the mapping contrast diagram through an image differential identification network, and clustering each differential pixel point to obtain each differential image; and identifying the image difference type of each difference image, and screening a target difference image of the target image difference type and point position information of each difference pixel point of the target difference image as target defect information of the circuit board to be tested. According to the scheme, the standard structure image of the standard circuit board and the structure image to be detected of the circuit board to be detected are subjected to mapping comparison, the image difference type of the difference image corresponding to the difference pixel points in the two structure images and the point position information of each difference pixel point are identified, so that the target defect information of the circuit board to be detected is determined, the circuit board to be detected does not need to be subjected to image identification directly, the defect detection efficiency is improved, the position information of the difference pixel points can be visually displayed through mapping comparison, and the situation that whether the difference pixel points are the difference pixel points or not through neural network analysis is avoided. Secondly, the defect information of the circuit board to be detected is identified and analyzed from the multi-angle standard structure image, the difference identification taking the pixel points as granularity and the difference image clustered by the different pixel points, so that the comprehensiveness, fineness and observability of identifying the defect information are improved, and the accuracy of detecting the surface defects of the PCBA is comprehensively improved under the condition of ensuring the identification efficiency by using a multi-angle pixel cluster comparison and identification strategy.
Drawings
FIG. 1 is a flow chart of a method for detecting surface defects of a PCBA in one embodiment;
FIG. 2 is a block diagram of a PCBA surface defect detection system in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The PCBA surface defect detection method provided by the embodiment of the application can be applied to the application environment of PBAC detection. The method can be applied to the terminal, the server and a system comprising the terminal and the server, and is realized through interaction of the terminal and the server. The terminal may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and the like. The terminal directly performs mapping comparison on the standard structure image of the standard circuit board and the structure image to be detected of the circuit board to be detected, and identifies the image difference type of the difference image corresponding to the difference pixel point in the two structure images and the point position information of each difference pixel point, so that the target defect information of the circuit board to be detected is determined, the circuit board to be detected does not need to be directly subjected to image identification, the defect detection efficiency is improved, the mapping comparison can visually display the position information of the difference pixel point, and the situation that whether the difference pixel point is the difference pixel point or not is analyzed through a neural network is avoided. Secondly, the defect information of the circuit board to be detected is identified and analyzed from the multi-angle standard structure image, the difference identification taking the pixel points as granularity and the difference image clustered by the different pixel points, so that the comprehensiveness, fineness and observability of identifying the defect information are improved, and the accuracy of detecting the surface defects of the PCBA is comprehensively improved under the condition of ensuring the identification efficiency by using a multi-angle pixel cluster comparison and identification strategy.
In one embodiment, as shown in fig. 1, a PCBA surface defect detection method is provided, and the method is applied to a terminal for illustration, and includes the following steps:
Step S101, obtaining standard structure images of all angles of a standard circuit board and to-be-detected structure images of all angles of a to-be-detected circuit board, and carrying out image mapping processing on the standard structure images of the angles and the to-be-detected structure images of the angles according to each angle to obtain an angle mapping comparison diagram.
In this embodiment, the terminal responds to the information uploading operation of the staff, acquires the circuit board which has no surface defect and is the same type, specification, manufacturer and production line as the circuit board to be tested, and acquires the standard structure images of all angles of the standard circuit board through the imaging equipment arranged at multiple angles. Then, the terminal is arranged at the same position of the standard circuit board through the equipment arranged at the multiple angles, namely, the structure images to be tested of the angles of the circuit boards to be tested are the structure shape images of different circuit boards, such as a surface structure diagram of the circuit board, a side surface structure diagram of the circuit board, a bottom surface structure diagram of the circuit board and the like. And then the terminal performs structural mapping processing on each structural image to be detected with the same angle and the standard structural image through a two-dimensional mapping technology to obtain a mapping contrast diagram of each angle of each circuit board to be detected. The specific mapping process will be described in detail later.
Step S102, identifying the difference pixel points in the mapping contrast diagram through an image difference identification network, and carrying out clustering processing on each difference pixel point to obtain each difference image.
In this embodiment, the terminal identifies, through the image difference identification network, difference pixel points in the mapping contrast map, and performs clustering processing on each difference pixel point to obtain each difference image. The difference image comprises different pixel points and pixel points in the surrounding range of the different pixel points. The image difference recognition network is a Faster-RCNN (Region-CNN) image detection neural network based on a deep learning neural network. The specific identification process will be described in detail later.
Step S103, identifying the image difference type of each difference image, and screening the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image as the target defect information of the circuit board to be tested.
In this embodiment, the terminal identifies the image difference type of each difference image, and screens the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image as the target defect information of the circuit board to be tested. The image difference types corresponding to the sub-structure information of the plate structures of different circuit boards are different, wherein the circuit boards comprise plate structures of areas where the components of a plurality of different component types are located, each plate structure comprises the sub-structure information of the area where the component is located, the image difference type corresponding to each sub-structure information is a surface defect type corresponding to each sub-structure information, and the surface defect types comprise, but are not limited to, missing welding, overflowing welding, misplacement, jack misplacement, cracking, foaming, layering, missing welding and the like. The specific screening process will be described in detail later.
Based on the scheme, the standard structure image of the standard circuit board and the structure image to be detected of the circuit board to be detected are directly mapped and compared, and the image difference type of the difference image corresponding to the difference pixel point in the two structure images and the point position information of each difference pixel point are identified, so that the target defect information of the circuit board to be detected is determined, the circuit board to be detected does not need to be directly identified, the defect detection efficiency is improved, the position information of the difference pixel point can be visually displayed by mapping comparison, and the situation that whether the difference pixel point is the difference pixel point or not is analyzed through a neural network is avoided. Secondly, the defect information of the circuit board to be detected is identified and analyzed from the multi-angle standard structure image, the difference identification taking the pixel points as granularity and the difference image clustered by the different pixel points, so that the comprehensiveness, fineness and observability of identifying the defect information are improved, and the accuracy of detecting the surface defects of the PCBA is comprehensively improved under the condition of ensuring the identification efficiency by using a multi-angle pixel cluster comparison and identification strategy.
Optionally, performing image mapping processing on the standard structural image of the angle and the structural image to be detected of the angle to obtain a mapping contrast diagram of the angle, including: carrying out equidistant edge marking treatment on each structural image to obtain each edge position point of each structural image; and carrying out overlapping mapping processing on each edge position point based on the edge identification information of each edge position point to obtain a mapping comparison chart.
In the embodiment, the terminal performs edge equidistant marking processing on each structural image to obtain each edge position point of each structural image; and carrying out overlapping mapping processing on each edge position point based on the edge identification information of each edge position point to obtain a mapping comparison chart. The equidistant marking processing is that each image edge of the structural image is marked equidistantly according to a preset fixed distance to obtain each edge position point of each image edge, wherein each edge position point corresponds to one piece of edge identification information, for example, left 1 (the first edge position point from top to bottom of the left image edge), upper 3 (the third edge position point from left to right of the upper image edge) and the like.
Based on the scheme, the accuracy of the mapping contrast map obtained by the overlapped mapping is improved through the overlapped mapping processing after the equidistant marking.
Optionally, before identifying the difference pixel point in the mapping contrast chart through the image difference identification network, the method further includes: gridding the mapping comparison image to obtain each sub-mapping image of the mapping comparison image; and carrying out pixel point marking processing on each pixel point in the sub-mapping image according to a pixel point marking strategy aiming at each sub-mapping image to obtain point identification information of each pixel point.
In this embodiment, the terminal performs gridding processing on the mapping contrast map to obtain each sub-mapping image of the mapping contrast map; for each sub-map image. The length of the side of the grid is the length preset by the staff on the terminal, and the length can be changed along with the adjustment of the staff. And carrying out pixel point marking processing on each pixel point in the sub-mapping image according to a pixel point marking strategy aiming at each sub-mapping image to obtain point identification information of each pixel point. The pixel point marking process is to sequentially mark each pixel point in the sub-map image from top to bottom and from front to back to obtain point identification information of each pixel point, wherein each point identification information is used for identifying the position information of the pixel point in the sub-map image.
Based on the scheme, after gridding processing, point marking processing is carried out on each pixel point, so that the recognition accuracy of each pixel point is improved, and the analysis accuracy of an image is not improved.
Optionally, identifying, by the image difference identification network, a difference pixel point in the mapping contrast map includes: inputting first pixel points of a standard structure image with the same point identification information and second pixel points of a structure image to be detected into an image difference identification network to obtain difference identification results between the first pixel points and the second pixel points; and taking the first pixel point and the second pixel point which are the difference of the difference recognition result as the difference pixel points in the mapping contrast diagram.
In this embodiment, the terminal inputs the first pixel point of the standard structure image with the same point identification information and the second pixel point of the structure image to be detected into the image difference identification network to obtain a difference identification result between each first pixel point and each second pixel point. The image difference recognition result is a convolutional neural network (Convolutional Neural Networks, CNN).
And then, the terminal uses the first pixel point and the second pixel point, the difference recognition result of which is the result difference, as the difference pixel point in the mapping contrast diagram.
Based on the scheme, the difference identification is carried out on each pixel point, so that the identified granularity is thinned, and the accuracy of identifying the difference pixel points is improved.
Optionally, clustering is performed on each difference pixel point to obtain each difference image, including: identifying pixel point distance values among different pixel points based on point identification information of the different pixel points and sub-mapping images of the different pixel points; and clustering each difference pixel point and each pixel point among the difference pixel points based on the pixel point distance value among the difference pixel points to obtain each difference image.
In this embodiment, the terminal identifies the pixel pitch value between the different pixel points based on the point identification information of each different pixel point and the sub-map image to which each different pixel point belongs. The pixel point interval value comprises a transverse pixel point interval value and a longitudinal pixel point interval value between two different pixel points. And then, the terminal performs clustering processing on each difference pixel point and each pixel point among the difference pixel points based on the pixel point distance value among the difference pixel points to obtain each difference image. The clustering processing is performed by dividing a pixel interval threshold value lower than a pixel interval threshold value preset at a terminal into the same difference image, and dividing pixels included between two difference images into the same difference image.
Based on the scheme, the identification accuracy of the global difference features corresponding to the difference pixel points is improved by carrying out clustering processing on the difference pixel points.
Optionally, before identifying the image difference type of each difference image, further includes: identifying each plate structure of the standard circuit board and the position range of each plate structure in the structural image, and collecting the structural range ratio of each sub-structural information of each plate structure and the structural position information of each sub-structural information; identifying a sub-position range of each sub-structure information in the structural range of the plate structure based on the structural range duty ratio of each sub-structure information and the structural position information of each sub-structure information; and identifying the image position range of each sub-map image in the structural image, and determining the sub-structure information corresponding to each sub-map image based on the sub-position range to which each image position range belongs.
In this embodiment, the terminal identifies each board structure of the standard circuit board and a position range of each board structure in the structural image, and collects a structural range ratio of each sub-structural information of each board structure and structural position information of each sub-structural information. The structure range duty ratio and the structure position information are stored in the database, and the terminal identifies the structure range duty ratio of each piece of sub-structure information and the structure position information of each piece of sub-structure information in the database.
The terminal identifies the sub-position range of each sub-structure information in the structural range of the plate structure based on the structural range duty ratio of each sub-structure information and the structural position information of each sub-structure information, then the terminal identifies the image position range of each sub-map image in the structural image, and determines the sub-structure information corresponding to each sub-map image based on the sub-position range to which each image position range belongs.
Based on the scheme, the accuracy of identifying the surface defect type corresponding to each sub-map image is improved by identifying the sub-structure information corresponding to each sub-map image.
Optionally, identifying an image difference type of each difference image includes: for each difference image, identifying each target sub-map image corresponding to the difference image, and identifying target sub-structure information corresponding to the difference image based on the sub-structure information corresponding to each target sub-map image; and inquiring the surface defect type of the target substructure information in the database, and identifying the image difference type of the difference image through the image identification network based on the surface defect type of the target substructure information.
In this embodiment, the terminal identifies each target sub-map image corresponding to the difference image for each difference image, and identifies target sub-structure information corresponding to the difference image based on sub-structure information corresponding to each target sub-map image. Then, the terminal queries the surface defect type of the target substructure information in the database, and identifies the image difference type of the difference image through the image identification network based on the surface defect type of the target substructure information. Wherein the image recognition network is a deep convolutional neural network VGG16 (Visual Geometry Group 16).
Based on the scheme, the image classification recognition is carried out on each difference image through the image recognition network, so that the accuracy of recognizing the image difference type of each difference image is improved.
Optionally, screening the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image as the target defect information of the circuit board to be tested, including: responding to the data uploading operation of the staff, acquiring a difference type of the target image, and screening the difference image of the difference type of the target image from the difference images to serve as the target difference image; identifying, for each target difference image, point location information for each difference pixel based on an image location range of the target difference image in the structural image and point identification information for each difference pixel in the target difference image; and taking the image difference type of each target difference image and the point position information of each difference pixel point of each target difference image as target defect information of the circuit board to be tested.
In this embodiment, the terminal responds to the data uploading operation of the staff to obtain the difference type of the target image, and screens the difference image of the difference type of the target image from the difference images as the target difference image. Then, the terminal identifies, for each target difference image, point position information of each difference pixel based on an image position range of the target difference image in the structural image and point identification information of each difference pixel in the target difference image. And finally, the terminal takes the image difference type of each target difference image and the point position information of each difference pixel point of each target difference image as target defect information of the circuit board to be tested.
Based on the scheme, after the difference images of the difference types of the target images are screened, the point position information of each difference pixel point is positioned based on the point identification information, so that the target defect information of the circuit board to be detected is obtained, and the comprehensiveness and the accuracy of identifying the target defect information of the circuit board to be detected are improved.
It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.
Based on the same inventive concept, the embodiment of the application also provides a PCBA surface defect detection system for realizing the PCBA surface defect detection method. The implementation of the solution provided by the system is similar to that described in the above method, so the specific limitations in the embodiments of the PCBA surface defect detection system or systems provided below may be referred to above as limitations of the PCBA surface defect detection method, and will not be described herein.
In one embodiment, as shown in FIG. 2, a PCBA surface defect detection system is provided, comprising: an acquisition module 210, an identification module 220, and a screening module 230, wherein:
The obtaining module 210 is configured to obtain standard structural images of angles of the standard circuit board and to obtain to-be-tested structural images of the angles of the to-be-tested circuit board, and perform image mapping processing on the standard structural images of the angles and the to-be-tested structural images of the angles for each angle to obtain a mapping comparison graph of the angles;
The identifying module 220 is configured to identify, through an image difference identifying network, difference pixel points in the map contrast map, and perform clustering processing on each of the difference pixel points to obtain each difference image;
The screening module 230 is configured to identify an image difference type of each difference image, and screen a target difference image of a target image difference type and point position information of each difference pixel point of the target difference image as target defect information of the circuit board to be tested.
Optionally, the acquiring module 210 is specifically configured to:
Performing equidistant edge marking processing on each structural image to obtain each edge position point of each structural image;
And carrying out overlapping mapping processing on each edge position point based on the edge identification information of each edge position point to obtain a mapping comparison chart.
Optionally, the system further comprises:
the mapping module is used for carrying out gridding treatment on the mapping comparison graph to obtain each sub-mapping image of the mapping comparison graph;
the marking module is used for carrying out pixel marking processing on each pixel point in each sub-mapping image through a pixel point marking strategy to obtain point identification information of each pixel point.
Optionally, the identifying module 220 is specifically configured to:
inputting a first pixel point of a standard structure image with the same point identification information and a second pixel point of a structure image to be detected into an image difference identification network to obtain a difference identification result between each first pixel point and each second pixel point;
and taking the first pixel point and the second pixel point which are the difference of the difference recognition result as the difference pixel points in the mapping contrast diagram.
Optionally, the identifying module 220 is specifically configured to:
identifying pixel point distance values between the different pixel points based on the point identification information of the different pixel points and the sub-mapping image to which the different pixel points belong;
And clustering each difference pixel point and each pixel point between each difference pixel point based on the pixel point distance value between each difference pixel point to obtain each difference image.
Optionally, the system further includes:
The acquisition module is used for identifying each plate structure of the standard circuit board and the position range of each plate structure in the structural image, and acquiring the structural range ratio of each sub-structural information of each plate structure and the structural position information of each sub-structural information;
A range identifying module, configured to identify, in a structural range of the board structure, a sub-position range of each piece of sub-structure information based on a structural range ratio of each piece of sub-structure information and structural position information of each piece of sub-structure information;
The determining module is used for identifying the image position range of each sub-map image in the structural image and determining the sub-structure information corresponding to each sub-map image based on the sub-position range to which each image position range belongs.
Optionally, the screening module 230 is specifically configured to:
For each difference image, identifying each target sub-map image corresponding to the difference image, and identifying target sub-structure information corresponding to the difference image based on sub-structure information corresponding to each target sub-map image;
and inquiring the surface defect type of the target substructure information in a database, and identifying the image difference type of the difference image through an image identification network based on the surface defect type of the target substructure information.
Optionally, the screening module 230 is specifically configured to:
Responding to the data uploading operation of the staff, acquiring a difference type of the target image, and screening the difference image of the difference type of the target image from the difference images to serve as the target difference image;
identifying, for each target difference image, point location information for each difference pixel point based on an image location range of the target difference image in the structural image and point identification information for each difference pixel point in the target difference image;
And taking the image difference type of each target difference image and the point position information of each difference pixel point of each target difference image as target defect information of the circuit board to be tested.
The above-described modules in the PCBA surface defect detection system may be implemented in whole or in part by software, hardware, and combinations thereof.
The user information (including but not limited to user equipment information, user personal information, etc.) and the data (including but not limited to data for analysis, stored data, presented data, etc.) related to the present application are information and data authorized by the user or sufficiently authorized by each party.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.
Claims (9)
1. A method for detecting surface defects of a PCBA, the method comprising:
obtaining standard structure images of all angles of a standard circuit board and structure images to be detected of all angles of the circuit board to be detected, and carrying out image mapping processing on the standard structure images of all angles and the structure images to be detected of all angles according to each angle to obtain a mapping comparison graph of all angles;
Identifying differential pixel points in the mapping contrast diagram through an image differential identification network, and clustering each differential pixel point to obtain each differential image;
And identifying the image difference type of each difference image, and screening a target difference image of the target image difference type and point position information of each difference pixel point of the target difference image as target defect information of the circuit board to be tested.
2. The method according to claim 1, wherein the performing image mapping processing on the standard structural image of the angle and the structural image to be measured of the angle to obtain a mapping contrast diagram of the angle includes:
Performing equidistant edge marking processing on each structural image to obtain each edge position point of each structural image;
And carrying out overlapping mapping processing on each edge position point based on the edge identification information of each edge position point to obtain a mapping comparison chart.
3. The method of claim 1, wherein the identifying, by the image difference identification network, the difference pixels in the map contrast map is preceded by:
Gridding the mapping comparison graph to obtain each sub-mapping image of the mapping comparison graph;
And carrying out pixel point marking processing on each pixel point in each sub-mapping image by a pixel point marking strategy according to each sub-mapping image to obtain point identification information of each pixel point.
4. A method according to claim 3, wherein said identifying, by the image difference identification network, the difference pixels in the map contrast map comprises:
inputting a first pixel point of a standard structure image with the same point identification information and a second pixel point of a structure image to be detected into an image difference identification network to obtain a difference identification result between each first pixel point and each second pixel point;
and taking the first pixel point and the second pixel point which are the difference of the difference recognition result as the difference pixel points in the mapping contrast diagram.
5. A method according to claim 3, wherein clustering each of the difference pixels to obtain each difference image includes:
identifying pixel point distance values between the different pixel points based on the point identification information of the different pixel points and the sub-mapping image to which the different pixel points belong;
And clustering each difference pixel point and each pixel point between each difference pixel point based on the pixel point distance value between each difference pixel point to obtain each difference image.
6. The method of claim 1, wherein prior to identifying the image difference type for each difference image, further comprising:
Identifying each plate structure of the standard circuit board and the position range of each plate structure in the structural image, and collecting the structural range ratio of each sub-structural information of each plate structure and the structural position information of each sub-structural information;
Identifying a sub-positional range of each piece of sub-structural information in a structural range of the plate structure based on a structural range ratio of each piece of sub-structural information and structural positional information of each piece of sub-structural information;
and identifying the image position range of each sub-map image in the structural image, and determining the sub-structure information corresponding to each sub-map image based on the sub-position range to which each image position range belongs.
7. The method of claim 6, wherein the identifying the image difference type for each difference image comprises:
For each difference image, identifying each target sub-map image corresponding to the difference image, and identifying target sub-structure information corresponding to the difference image based on sub-structure information corresponding to each target sub-map image;
and inquiring the surface defect type of the target substructure information in a database, and identifying the image difference type of the difference image through an image identification network based on the surface defect type of the target substructure information.
8. The method according to claim 3, wherein the screening the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image as the target defect information of the circuit board to be tested includes:
Responding to the data uploading operation of the staff, acquiring a difference type of the target image, and screening the difference image of the difference type of the target image from the difference images to serve as the target difference image;
identifying, for each target difference image, point location information for each difference pixel point based on an image location range of the target difference image in the structural image and point identification information for each difference pixel point in the target difference image;
And taking the image difference type of each target difference image and the point position information of each difference pixel point of each target difference image as target defect information of the circuit board to be tested.
9. A PCBA surface defect detection system, the system comprising:
The acquisition module is used for acquiring standard structure images of all angles of the standard circuit board and to-be-detected structure images of all angles of the circuit board to be detected, and carrying out image mapping processing on the standard structure images of the angles and the to-be-detected structure images of the angles aiming at each angle to obtain a mapping comparison graph of the angles;
The identification module is used for identifying the difference pixel points in the mapping contrast diagram through an image difference identification network, and carrying out clustering treatment on each difference pixel point to obtain each difference image;
The screening module is used for identifying the image difference type of each difference image, screening the target difference image of the target image difference type and the point position information of each difference pixel point of the target difference image, and taking the point position information as the target defect information of the circuit board to be tested.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410446300.6A CN118052816A (en) | 2024-04-15 | 2024-04-15 | PCBA surface defect detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410446300.6A CN118052816A (en) | 2024-04-15 | 2024-04-15 | PCBA surface defect detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118052816A true CN118052816A (en) | 2024-05-17 |
Family
ID=91046891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410446300.6A Pending CN118052816A (en) | 2024-04-15 | 2024-04-15 | PCBA surface defect detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118052816A (en) |
-
2024
- 2024-04-15 CN CN202410446300.6A patent/CN118052816A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Akagic et al. | Pothole detection: An efficient vision based method using rgb color space image segmentation | |
| CN113313712B (en) | Battery gluing defect detection method and device, electronic equipment and storage medium | |
| CN116503415B (en) | Defect detection method, defect detection device, computer equipment and computer readable storage medium | |
| Chen et al. | A deep region-based pyramid neural network for automatic detection and multi-classification of various surface defects of aluminum alloys | |
| CN115035081B (en) | Industrial CT-based metal internal defect dangerous source positioning method and system | |
| CN115115825B (en) | Method, device, computer equipment and storage medium for detecting object in image | |
| CN118052816A (en) | PCBA surface defect detection method and system | |
| CN114913350B (en) | Material duplicate checking method, device, equipment and storage medium | |
| CN116051959A (en) | Target detection method and device | |
| CN115063473A (en) | Object height detection method and device, computer equipment and storage medium | |
| US20220245782A1 (en) | Method and apparatus for classifying image of displaying base plate | |
| CN116206125A (en) | Appearance defect identification method, appearance defect identification device, computer equipment and storage medium | |
| CN112084364A (en) | Object analysis method, local image search method, device, and storage medium | |
| Ashraf et al. | Efficient Pavement Crack Detection and Classification Using Custom YOLOv7 Model | |
| US11989908B2 (en) | Visual positioning method, mobile machine using the same, and computer readable storage medium | |
| CN116665177B (en) | Data processing method, device, electronic device and storage medium | |
| Ashraf et al. | Machine learning-based pavement crack detection, classification, and characterization: a review | |
| CN117437602B (en) | Dual-layer data calibration method, device, equipment and readable storage medium | |
| CN116522261B (en) | Risk information monitoring method and system based on big data | |
| CN114882058B (en) | Corner detection method, corner detection device and calibration plate | |
| CN111553418B (en) | Method and device for detecting neuron reconstruction errors and computer equipment | |
| CN117351859B (en) | Detection method, device and system for display module | |
| CN115063432B (en) | Method and device for identifying alignment mark pattern, computer equipment and storage medium | |
| Li et al. | OUR-Net: A Multi-Frequency Network With Octave Max Unpooling and Octave Convolution Residual Block for Pavement Crack Segmentation | |
| CN116071418A (en) | Method, apparatus, computer device, storage medium and computer program product for measuring build-up dimensions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |