Circuit board defect detection system and method

Abstract

The present invention provides a circuit board defect detection system and method, which respectively perform machine vision photography and identification on the circuit board under a first lighting condition and a second lighting condition, obtain first and second alignment mark feature information on the surface of the circuit board and defect state information of a wafer part and wiring state feature information, and can accurately differentiate and identify defects existing in the wafer and wiring through different lighting conditions, use the first and second alignment mark feature information to perform a same area orientation comparison on the defect state information of the wafer part and the wiring state feature information, associate and identify defects existing in the wafer part and the wiring part of the circuit board, thereby identifying defective bad point information that affects the normal working performance of the circuit board, providing a reliable basis for judging the quality state of the circuit board, and performing integrated and diversified defect detection on the circuit board by identifying defects existing on the wafer and wiring solder joints of the circuit board, thereby realizing reliable performance quality screening of the circuit board.

Description

Technical Field

[0001] The present invention relates to the field of circuit board processing, and in particular to a circuit board defect detection system and method. Background Art

[0002] As a precision electronic device, the circuit board has a large number of components soldered on the surface of the wafer and the corresponding components are wired and connected. In the actual production process of the circuit board, it is inevitable that there will be defects in the surface morphology of the wafer, defects in component soldering, and defects in wiring connections. These defects will more or less affect the working performance of the circuit board. When the number of defects is large enough or the defects are formed at the key position of the circuit board, the circuit board will become a defective product and cannot achieve the computing performance corresponding to the original design. Existing circuit board defect detection is to identify wiring defects on the surface of the circuit board. It is impossible to carry out detailed correlation detection of different types of defects on the surface of the circuit board, it is impossible to locate all defects on the surface of the circuit board and accurately identify the quality status of the circuit board, and it is impossible to reliably screen the performance quality of the circuit board. Summary of the invention

[0003] The purpose of the present invention is to provide a circuit board defect detection system and method, which respectively perform machine vision photography and identification on the circuit board under a first lighting condition and a second lighting condition, obtain first and second alignment mark feature information on the surface of the circuit board and defect status information of the wafer part and wiring status feature information, and can accurately differentiate and identify defects existing in the wafer and wiring through different lighting conditions, use the first and second alignment mark feature information to compare the defect status information of the wafer part and the wiring status feature information in the same area, associate and identify defects existing in the wafer part and the wiring part of the circuit board, thereby identifying defective bad point information that affects the normal working performance of the circuit board, providing a reliable basis for judging the quality status of the circuit board, and performing integrated and diversified defect detection on the circuit board by identifying defects on the wafer and wiring solder joints of the circuit board, thereby realizing reliable performance quality screening of the circuit board.

[0004] The present invention is achieved through the following technical solutions:

[0005] A circuit board defect detection system, comprising:

[0006] A first visual shooting module, used for performing machine vision shooting of the circuit board under a first lighting condition to obtain a first circuit board image;

[0007] A first image recognition module, configured to recognize the first circuit board image based on the lighting characteristic information of the first lighting condition, and obtain first alignment mark characteristic information on the surface of the circuit board and defect status information of a wafer portion;

[0008] A second visual shooting module, used for performing machine vision shooting of the circuit board under a second lighting condition to obtain a second circuit board image;

[0009] A second image recognition module, configured to recognize the second circuit board image based on the lighting characteristic information of the second lighting condition, and obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board;

[0010] A wafer and wiring relative relationship determination module, used to compare the defect state information of the wafer part and the wiring state characteristic information in the same area based on the first alignment mark characteristic information and the second alignment mark characteristic information, and determine the relative relationship state between the wafer part and the wiring part of the circuit board;

[0011] The circuit board quality determination module is used to identify the existence of defective bad pixel information on the surface of the circuit board based on the relative relationship state, and determine the quality state of the circuit board based on the defective bad pixel information.

[0012] Optionally, the first visual shooting module is used to perform machine vision shooting of the circuit board under a first lighting condition to obtain a first circuit board image, including:

[0013] Under the condition of uniform surface light beam illumination with a single wavelength, the surface of the circuit board is photographed by machine vision in different regions to obtain a plurality of first images of different regions of the surface of the circuit board; based on the shooting orientation information of the machine vision regional photography, all the first images are spliced ​​to obtain a first circuit board image;

[0014] The first image recognition module is used to recognize the first circuit board image based on the lighting characteristic information of the first lighting condition, and obtain the first alignment mark characteristic information on the surface of the circuit board and the defect state information of the wafer part, including:

[0015] Based on the light wavelength corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing that matches the light wavelength to obtain a corresponding circuit board image; the circuit board image is subjected to pixel contour recognition processing to obtain first edge contour feature information of the alignment mark on the surface of the circuit board and edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board.

[0016] Optionally, the second visual shooting module is used to perform machine vision shooting of the circuit board under a second lighting condition to obtain a second circuit board image, including:

[0017] Performing machine vision scanning and photographing of the circuit board surface under white structured light illumination with a pattern of alternating light and dark stripes to obtain a second circuit board image corresponding to a global optical pattern formed by the circuit board surface after the white structured light is reflected;

[0018] The second image recognition module is used to recognize the second circuit board image based on the lighting characteristic information of the second lighting condition to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board, including:

[0019] After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board.

[0020] Optionally, the wafer and wiring relative relationship determination module is used to compare the defect state information of the wafer part and the wiring state characteristic information in the same area based on the first alignment mark characteristic information and the second alignment mark characteristic information, and determine the relative relationship state between the wafer part and the wiring part of the circuit board, including:

[0021] Comparing the first edge profile feature information and the second edge profile feature information, the orientation of the comparison mark is compared to determine the orientation deviation between the two;

[0022] Based on the orientation deviation, the edge contour feature information of the morphology defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board are adjusted with respect to the projection orientation on the two-dimensional plane, so as to project the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect into the corresponding two-dimensional plane space, thereby performing the orientation comparison of the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect in the same area, and obtaining the relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board;

[0023] The circuit board quality determination module is used to identify defective bad pixel information on the surface of the circuit board based on the relative relationship state, and determine the quality state of the circuit board based on the defective bad pixel information, including:

[0024] Based on the relative distance state, the wafer defect-wiring defect / solder point defect pairs that affect the normal operation of the circuit board are identified from all wafer defects and all wiring defects / solder point defects of the circuit board, and are used as defective bad spots on the surface of the circuit board; and based on the distribution density of defective bad spots corresponding to the surface of the circuit board, it is determined whether the circuit board is a good circuit board or a defective circuit board.

[0025] Optionally, a shooting positioning module is further included, which is used to position the shooting of the first visual shooting module and the second visual shooting module, so as to ensure that the shooting positions of the first visual shooting module and the second visual shooting module are consistent, which includes:

[0026] Step S1: There are four vertex positioning points on the shooting table where the circuit board is placed, and each positioning point can be recognized by the first visual shooting module and the second visual shooting module. The following formula (1) is used to determine whether the captured circuit board image has image distortion according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the second visual shooting module.

[0027] (1),

[0028] In the above formula (1), Indicates the judgment value for judging whether the captured circuit board image has image distortion; Indicates the coordinates of the upper left corner positioning point in the circuit board image captured by the camera module; Indicates the coordinates of the upper right corner positioning point in the circuit board image captured by the camera module; Indicates the coordinates of the lower left corner positioning point in the circuit board image captured by the camera module; Indicates the coordinates of the lower right corner positioning point in the circuit board image captured by the camera module; || indicates parallel to; && indicates a logical relationship and; Represents a judgment function. If the formula in the brackets is true, the function value is 1. If the formula in the brackets is not true, the function value is 0.

[0029] like , it means that there is no image distortion in the captured circuit board image;

[0030] like , it means that the captured circuit board image has image distortion;

[0031] If there is image distortion, the shooting module is adjusted and the above step S1 is repeated until the circuit board images shot by the first visual shooting module and the second visual shooting module have no image distortion;

[0032] Step S2: when there is no image distortion in the circuit board images captured by the first visual shooting module and the second visual shooting module, the shooting parameters of the first visual shooting module and the second visual shooting module are set to be consistent, and then the following formula (2) is used to determine whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module.

[0033] (2)

[0034] In the above formula (2), A determination value indicating whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent; Indicates the first circuit board image captured by the first visual shooting module The coordinates of the vertex positioning points; Indicates the first image of the circuit board captured by the second visual shooting module The coordinates of the vertex positioning points;

[0035] like , it means that the shooting positions of the first visual shooting module and the second visual shooting module are consistent;

[0036] like , it means that the shooting positions of the first visual shooting module and the second visual shooting module are inconsistent;

[0037] Step S3, when the shooting positions of the first visual shooting module and the second visual shooting module are inconsistent, the following formula (3) is used to control the direction of the up and down movement adjustment of the second visual shooting module according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module.

[0038] (3)

[0039] In the above formula (3), R represents the direction control value of the up and down movement adjustment of the second visual shooting module; It means to find the distance between the two coordinate points in the brackets;

[0040] like , it means controlling the second visual shooting module to move upward and repeating the above steps S2 and S3 until the shooting positions of the first visual shooting module and the second visual shooting module are consistent;

[0041] like , it means controlling the second visual shooting module to move downward and repeating the above steps S2 and S3 until the shooting positions of the first visual shooting module and the second visual shooting module are consistent.

[0042] A circuit board defect detection method, comprising:

[0043] Performing machine vision photography of the circuit board under a first lighting condition to obtain a first circuit board image; recognizing the first circuit board image based on lighting feature information of the first lighting condition to obtain first alignment mark feature information on the surface of the circuit board and defect status information of a wafer portion;

[0044] Performing machine vision photography on the circuit board under a second lighting condition to obtain a second circuit board image; recognizing the second circuit board image based on lighting characteristic information of the second lighting condition to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board;

[0045] Based on the first alignment mark characteristic information and the second alignment mark characteristic information, the defect status information of the wafer part and the wiring status characteristic information are compared in the same area orientation to determine the relative relationship status between the wafer part and the wiring part of the circuit board; based on the relative relationship status, the defective bad point information on the surface of the circuit board is identified, and based on the defective bad point information, the quality status of the circuit board is determined.

[0046] Optionally, a machine vision photograph is performed on the circuit board under a first lighting condition to obtain a first circuit board image; based on lighting feature information of the first lighting condition, the first circuit board image is recognized to obtain first alignment mark feature information on the surface of the circuit board and defect status information of a wafer part, including:

[0047] Under the condition of uniform surface light beam illumination with a single wavelength, the surface of the circuit board is photographed by machine vision in different regions to obtain a plurality of first images of different regions of the surface of the circuit board; based on the shooting orientation information of the machine vision regional photography, all the first images are spliced ​​to obtain a first circuit board image;

[0048] Based on the light wavelength corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing that matches the light wavelength to obtain a corresponding circuit board image; the circuit board image is subjected to pixel contour recognition processing to obtain first edge contour feature information of the alignment mark on the surface of the circuit board and edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board.

[0049] Optionally, the circuit board is photographed by machine vision under a second lighting condition to obtain a second circuit board image; based on the lighting characteristic information of the second lighting condition, the second circuit board image is recognized to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board, including:

[0050] Performing machine vision scanning and photographing of the circuit board surface under white structured light illumination with a pattern of alternating light and dark stripes to obtain a second circuit board image corresponding to a global optical pattern formed by the circuit board surface after the white structured light is reflected;

[0051] After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board.

[0052] Optionally, based on the first alignment mark characteristic information and the second alignment mark characteristic information, the defect state information of the wafer part and the wiring state characteristic information are compared in the same area to determine the relative relationship state between the wafer part and the wiring part of the circuit board; based on the relative relationship state, the defect bad point information on the surface of the circuit board is identified, and based on the defect bad point information, the quality state of the circuit board is determined, including:

[0053] Comparing the first edge profile feature information and the second edge profile feature information, the orientation of the comparison mark is compared to determine the orientation deviation between the two;

[0054] Based on the orientation deviation, the edge contour feature information of the morphology defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board are adjusted with respect to the projection orientation on the two-dimensional plane, so as to project the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect into the corresponding two-dimensional plane space, thereby performing the orientation comparison of the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect in the same area, and obtaining the relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board;

[0055] Based on the relative distance state, the wafer defect-wiring defect / solder point defect pairs that affect the normal operation of the circuit board are identified from all wafer defects and all wiring defects / solder point defects of the circuit board, and are used as defective bad spots on the surface of the circuit board; and based on the distribution density of defective bad spots corresponding to the surface of the circuit board, it is determined whether the circuit board is a good circuit board or a defective circuit board.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The circuit board defect detection system and method provided by the present application respectively perform machine vision photography and identification of the circuit board under a first lighting condition and a second lighting condition, obtain first and second alignment mark feature information on the surface of the circuit board and defect status information of the wafer part and wiring status feature information, and can accurately differentiate and identify defects existing in the wafer and wiring through different lighting conditions, use the first and second alignment mark feature information to compare the defect status information of the wafer part and the wiring status feature information in the same area, associate and identify defects existing in the wafer part and the wiring part of the circuit board, thereby identifying defective bad point information that affects the normal working performance of the circuit board, providing a reliable basis for judging the quality status of the circuit board, and performing integrated and diversified defect detection on the circuit board by identifying defects on the wafer and wiring solder joints of the circuit board, thereby realizing reliable performance quality screening of the circuit board. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0059] Figure 1 A schematic structural diagram of a circuit board defect detection system provided by the present invention.

[0060] Figure 2 A schematic diagram of a flow chart of a circuit board defect detection method provided by the present invention. DETAILED DESCRIPTION

[0061] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some structures related to the present application are shown in the accompanying drawings, rather than all structures. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0062] The terms "including" and "having" and any variations thereof in this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0063] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0064] See also Figure 1 As shown, an embodiment of the present application provides a circuit board defect detection system. The circuit board defect detection system includes:

[0065] A first visual shooting module, used for performing machine vision shooting of the circuit board under a first lighting condition to obtain a first circuit board image;

[0066] A first image recognition module is used to recognize the first circuit board image based on the lighting characteristic information of the first lighting condition, and obtain first alignment mark characteristic information on the surface of the circuit board and defect status information of the wafer part;

[0067] A second visual shooting module is used to perform machine vision shooting of the circuit board under a second lighting condition to obtain a second circuit board image;

[0068] A second image recognition module is used to recognize the second circuit board image based on the lighting characteristic information of the second lighting condition, and obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board;

[0069] A wafer and wiring relative relationship determination module, used to compare the defect state information of the wafer part and the wiring state characteristic information in the same area based on the first alignment mark characteristic information and the second alignment mark characteristic information, and determine the relative relationship state between the wafer part and the wiring part of the circuit board;

[0070] The circuit board quality determination module is used to identify the existence of defective bad pixel information on the surface of the circuit board based on the relative relationship state, and determine the quality state of the circuit board based on the defective bad pixel information.

[0071] The beneficial effects of the above embodiments are as follows: the circuit board defect detection system performs machine vision photography and identification on the circuit board under the first lighting condition and the second lighting condition respectively, obtains the first and second alignment mark feature information on the surface of the circuit board and the defect status information of the wafer part and the wiring status feature information, and can accurately differentiate and identify the defects existing in the wafer and wiring through different lighting conditions, and uses the first and second alignment mark feature information to compare the defect status information of the wafer part and the wiring status feature information in the same area, and associates and identifies the defects existing in the wafer part and the wiring part of the circuit board, thereby identifying the defect bad point information that affects the normal working performance of the circuit board, providing a reliable basis for judging the quality status of the circuit board, and performing integrated and diversified defect detection on the circuit board by identifying the defects existing on the wafer and wiring solder joints of the circuit board, thereby realizing reliable performance quality screening of the circuit board.

[0072] In another embodiment, the first visual shooting module is used to perform machine vision shooting of the circuit board under the first lighting condition to obtain a first circuit board image, including:

[0073] Under the condition of uniform surface light beam illumination with a single wavelength, the surface of the circuit board is photographed by machine vision in different regions to obtain a plurality of first images of different regions on the surface of the circuit board; based on the shooting orientation information of the machine vision regional photography, all the first images are spliced ​​to obtain a first circuit board image;

[0074] The first image recognition module is used to recognize the first circuit board image based on the lighting characteristic information of the first lighting condition, and obtain the first alignment mark characteristic information on the surface of the circuit board and the defect state information of the wafer part, including:

[0075] Based on the wavelength of light corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing that matches the wavelength of light to obtain a corresponding circuit board image; the circuit board image is subjected to pixel contour recognition processing to obtain first edge contour feature information of the alignment mark on the surface of the circuit board and edge contour feature information of the morphological defects of the wafer part on the surface of the circuit board.

[0076] The beneficial effects of the above embodiments are that the circuit board includes a wafer and a circuit wiring arranged on the surface of the wafer. The wafer and the circuit wiring inevitably have defects in the manufacturing process during the manufacturing process, resulting in structural defects in the wafer and the circuit wiring. These structural defects will have different degrees of impact on the performance of the circuit board. For example, when the wafer defect is located outside the circuit wiring area, it will not have an adverse effect on the performance of the circuit board; when the circuit wiring defect is located in the key part of the circuit, it may cause the entire circuit board to be scrapped. In addition, in order to accurately install the circuit board, an alignment mark will be formed on the surface of the circuit board. The alignment mark can be but not limited to a cross mark. The alignment mark can identify and calibrate the orientation of the circuit board (i.e., identify and calibrate the left and right front and back directions of the circuit board). In addition, the defect characteristics of the wafer and the circuit wiring on the circuit board are not the same. The wafer defect is mainly a defect in the surface morphology of the wafer, and the circuit wiring defect is mainly a defect in the connection and / or direction of the circuit wiring. If the wafer and circuit wiring defects are identified at the same time in the same machine vision recognition, it may not be possible to accurately distinguish and locate different types of defects, and it is impossible to comprehensively and accurately identify all types of defects on the surface of the circuit board. To this end, by performing machine vision recognition on the circuit board under different lighting conditions, the wafer defects and circuit wiring defects can be differentiated and directional recognized. Accordingly, the circuit board surface is photographed by machine vision in different regions under the illumination condition of a uniform surface beam with a single wavelength, and several first images of different regions of the circuit board surface are obtained, wherein the uniform surface beam can be, but not limited to, a beam of red or green wavelength, so that the uniform surface beam can perform uniform illumination on the corresponding region of the circuit board surface, so that the illuminated circuit board surface region can reflect the surface beam, and the reflected beam contains the surface morphology structure details of the wafer region corresponding to the circuit board surface; based on the shooting orientation information of the machine vision regional shooting, all the first images are spliced ​​to obtain the first circuit board image, thereby realizing the global wafer structure visual representation of the circuit board surface. Then, based on the light wavelength corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing matching the light wavelength to obtain the corresponding circuit board image, wherein the image layer extraction processing is performed by extracting the spectral data of the corresponding wavelength of the first circuit board image, which belongs to the conventional technical means in the field and is not described in detail here. The circuit board image is then processed with pixel contour recognition to obtain the first edge contour feature information of the alignment mark on the circuit board surface and the edge contour feature information of the morphological defect of the wafer part on the circuit board surface. This enables the alignment mark on the circuit board surface and the morphological defect on the wafer surface to be comprehensively and accurately identified.

[0077] In another embodiment, the second visual shooting module is used to perform machine vision shooting of the circuit board under a second lighting condition to obtain a second circuit board image, including:

[0078] Performing machine vision scanning and photographing of the circuit board surface under white structured light illumination with a pattern of alternating light and dark stripes to obtain a second circuit board image corresponding to a global optical pattern formed by the circuit board surface after the white structured light is reflected;

[0079] The second image recognition module is used to recognize the second circuit board image based on the lighting characteristic information of the second lighting condition to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board, including:

[0080] After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board.

[0081] The beneficial effects of the above embodiments are that the circuit wiring defects on the surface of the circuit board mainly include wiring defects and solder joint defects, and the defect shapes they present are rich and varied. In order to comprehensively identify the circuit wiring defects, the surface of the circuit board is scanned and photographed by machine vision under the illumination conditions of white structured light with a pattern of alternating light and dark stripes, and a second circuit board image corresponding to the global optical pattern formed after the circuit board surface reflects the white structured light is obtained, that is, the circuit board surface is illuminated by structured light with a periodic pattern of alternating light and dark stripes formed by a white light source, so that the circuit board surface will adjust the light intensity of the structured light and reflect it to form a corresponding global optical pattern, so that the second circuit board image can comprehensively capture the global optical pattern. After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defects / solder point defects on the surface of the circuit board. The above-mentioned comparison processing is a common means of structured light detection and will not be described in detail here. The above-mentioned method can accurately locate and identify the wiring defects / solder point defects on the surface of the circuit board.

[0082] In another embodiment, the wafer-to-wiring relative relationship determination module is used to compare the defect state information of the wafer portion and the wiring state characteristic information in the same area based on the first alignment mark characteristic information and the second alignment mark characteristic information, and determine the relative relationship state between the wafer portion and the wiring portion of the circuit board, including:

[0083] Comparing the orientation of the first edge profile feature information and the second edge profile feature information to determine the orientation deviation between the two;

[0084] Based on the orientation deviation, the edge contour feature information of the morphology defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board are adjusted in terms of projection orientation on a two-dimensional plane, so as to project the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect to a corresponding two-dimensional plane space, thereby performing an orientation comparison of the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect in the same area, and obtaining a relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board;

[0085] The circuit board quality determination module is used to identify defective bad pixel information on the surface of the circuit board based on the relative relationship state, and determine the quality state of the circuit board based on the defective bad pixel information, including:

[0086] Based on the relative distance state, the wafer defect-wiring defect / solder point defect pairs that affect the normal operation of the circuit board are identified from all wafer defects and all wiring defects / solder point defects of the circuit board, and are used as defective bad spots on the surface of the circuit board; then, based on the corresponding defective bad spot distribution density on the surface of the circuit board, it is determined whether the circuit board is a good circuit board or a defective circuit board.

[0087] The beneficial effect of the above embodiment is that the first edge contour feature information and the second edge contour feature information are both edge contour recognition results of the same alignment mark on the circuit board, but the placement orientation of the circuit board is not necessarily exactly the same when the machine vision shooting is performed under the first lighting condition and the second lighting condition, which makes the edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board have orientation differences. In order to facilitate the comparison of the edge contour feature information of the morphological defect of the wafer part and the edge contour feature information of the wiring defect / solder point defect in the same orientation space Analysis, first compare the orientation of the first edge profile feature information and the second edge profile feature information to determine the orientation deviation between the two, and then adjust the projection orientation of the morphological defect edge profile feature information of the wafer part on the surface of the circuit board and the wiring defect / solder point defect edge profile feature information on the surface of the circuit board on the two-dimensional plane based on the orientation deviation, so as to project the morphological defect edge profile feature information and the wiring defect / solder point defect edge profile feature information to the corresponding two-dimensional plane space, so as to obtain the relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board. Furthermore, the relative distance between the wafer defect and the wiring defect / solder point defect of the circuit board is compared with the preset distance threshold. If the relative distance is less than or equal to the preset distance threshold, the corresponding wafer defect and all wiring defects / solder point defects are determined as wafer defects-wiring defects / solder point defects that affect the normal operation of the circuit board, and used as defective bad spots on the surface of the circuit board. The defective bad pixel distribution density corresponding to the surface of the circuit board (i.e., the number of defective bad pixel distribution per unit area) is then compared with the preset density threshold. If the defective bad pixel distribution density is greater than the preset density threshold, the circuit board is determined to be a defective circuit board. Otherwise, the circuit board is determined to be a good circuit board, thereby quantitatively distinguishing the quality status of the circuit board.

[0088] In another embodiment, a shooting positioning module is further included, which is used to position the shooting of the first visual shooting module and the second visual shooting module, so as to ensure that the shooting positions of the first visual shooting module and the second visual shooting module are consistent, and includes:

[0089] Step S1: There are four vertex positioning points on the shooting table where the circuit board is placed, and each positioning point can be recognized by the first visual shooting module and the second visual shooting module. The following formula (1) is used to determine whether the captured circuit board image has image distortion according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the second visual shooting module.

[0090] (1),

[0091] In the above formula (1), Indicates the judgment value for judging whether the captured circuit board image has image distortion; Indicates the coordinates of the upper left corner positioning point in the circuit board image captured by the camera module; Indicates the coordinates of the upper right corner positioning point in the circuit board image captured by the camera module; Indicates the coordinates of the lower left corner positioning point in the circuit board image captured by the camera module; Indicates the coordinates of the lower right corner positioning point in the circuit board image captured by the camera module; || indicates parallel to; && indicates a logical relationship and; Represents a judgment function. If the formula in the brackets is true, the function value is 1. If the formula in the brackets is not true, the function value is 0.

[0092] like , it means that there is no image distortion in the captured circuit board image;

[0093] like , it means that the captured circuit board image has image distortion;

[0094] If there is image distortion, the shooting module is adjusted and the above step S1 is repeated until the circuit board images shot by the first visual shooting module and the second visual shooting module have no image distortion;

[0095] Step S2: when there is no image distortion in the circuit board images captured by the first visual shooting module and the second visual shooting module, the shooting parameters of the first visual shooting module and the second visual shooting module are set to be consistent, and then the following formula (2) is used to determine whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module.

[0096] (2)

[0097] In the above formula (2), A determination value indicating whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent; Indicates the first circuit board image captured by the first visual shooting module The coordinates of the vertex positioning points; Indicates the first image of the circuit board captured by the second visual shooting module The coordinates of the vertex positioning points;

[0098] like , it means that the shooting positions of the first visual shooting module and the second visual shooting module are consistent;

[0099] like , it means that the shooting positions of the first visual shooting module and the second visual shooting module are inconsistent;

[0100] Step S3, when the shooting positions of the first visual shooting module and the second visual shooting module are inconsistent, the following formula (3) is used to control the direction of the up and down movement adjustment of the second visual shooting module according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module.

[0101] (3)

[0102] In the above formula (3), R represents the direction control value of the up and down movement adjustment of the second visual shooting module; It means to find the distance between the two coordinate points in the brackets;

[0103] like , it means controlling the second visual shooting module to move upward and repeating the above steps S2 and S3 until the shooting positions of the first visual shooting module and the second visual shooting module are consistent;

[0104] like , it means controlling the second visual shooting module to move downward and repeating the above steps S2 and S3 until the shooting positions of the first visual shooting module and the second visual shooting module are consistent.

[0105] The beneficial effect of the above embodiment is that, by using the above formula (1), according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the second visual shooting module, it is judged whether the captured circuit board image has image distortion, thereby avoiding the difference in shooting positioning caused by image distortion; then by using the above formula (2), according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module, it is judged whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent, thereby intelligent judgment reflects the intelligence of the system; finally, by using the above formula (3), according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module, the direction of the up and down movement adjustment of the second visual shooting module is controlled, thereby ensuring the consistency of the position in the up and down directions and ensuring the reliability of the system.

[0106] See also Figure 2 As shown, a circuit board defect detection method provided by an embodiment of the present application. The circuit board defect detection method includes:

[0107] Performing machine vision photography of the circuit board under a first lighting condition to obtain a first circuit board image; based on lighting characteristic information of the first lighting condition, identifying the first circuit board image to obtain first alignment mark characteristic information on the surface of the circuit board and defect status information of a wafer portion;

[0108] Performing machine vision photography on the circuit board under a second lighting condition to obtain a second circuit board image; based on lighting characteristic information of the second lighting condition, recognizing the second circuit board image to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board;

[0109] Based on the first alignment mark characteristic information and the second alignment mark characteristic information, the defect status information of the wafer part and the wiring status characteristic information are compared in the same area orientation to determine the relative relationship status between the wafer part and the wiring part of the circuit board; based on the relative relationship status, the defect bad point information on the surface of the circuit board is identified, and based on the defect bad point information, the quality status of the circuit board is determined.

[0110] The beneficial effects of the above embodiments are as follows: the circuit board defect detection method performs machine vision photography and identification on the circuit board under the first lighting condition and the second lighting condition respectively, obtains the first and second alignment mark feature information on the surface of the circuit board and the defect status information of the wafer part and the wiring status feature information, and can accurately differentiate and identify the defects existing in the wafer and wiring through different lighting conditions, uses the first and second alignment mark feature information to compare the defect status information of the wafer part and the wiring status feature information in the same area, associates and identifies the defects existing in the wafer part and the wiring part of the circuit board, thereby identifying the defect bad point information that affects the normal working performance of the circuit board, and providing a reliable basis for judging the quality status of the circuit board, and performs integrated and diversified defect detection on the circuit board by identifying the defects existing on the wafer and wiring solder joints of the circuit board, thereby realizing reliable performance quality screening of the circuit board.

[0111] In another embodiment, a circuit board is photographed by machine vision under a first lighting condition to obtain a first circuit board image; based on lighting feature information of the first lighting condition, the first circuit board image is recognized to obtain first alignment mark feature information on the surface of the circuit board and defect status information of a wafer portion, including:

[0112] Under the condition of uniform surface light beam illumination with a single wavelength, the surface of the circuit board is photographed by machine vision in different regions to obtain a plurality of first images of different regions on the surface of the circuit board; based on the shooting orientation information of the machine vision regional photography, all the first images are spliced ​​to obtain a first circuit board image;

[0113] Based on the wavelength of light corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing that matches the wavelength of light to obtain a corresponding circuit board image; the circuit board image is subjected to pixel contour recognition processing to obtain first edge contour feature information of the alignment mark on the surface of the circuit board and edge contour feature information of the morphological defects of the wafer part on the surface of the circuit board.

[0114] The beneficial effects of the above embodiments are that the circuit board includes a wafer and a circuit wiring arranged on the surface of the wafer. The wafer and the circuit wiring inevitably have defects in the manufacturing process during the manufacturing process, resulting in structural defects in the wafer and the circuit wiring. These structural defects will have different degrees of impact on the performance of the circuit board. For example, when the wafer defect is located outside the circuit wiring area, it will not have an adverse effect on the performance of the circuit board; when the circuit wiring defect is located in the key part of the circuit, it may cause the entire circuit board to be scrapped. In addition, in order to accurately install the circuit board, an alignment mark will be formed on the surface of the circuit board. The alignment mark can be but not limited to a cross mark. The alignment mark can identify and calibrate the orientation of the circuit board (i.e., identify and calibrate the left and right front and back directions of the circuit board). In addition, the defect characteristics of the wafer and the circuit wiring on the circuit board are not the same. The wafer defect is mainly a defect in the surface morphology of the wafer, and the circuit wiring defect is mainly a defect in the connection and / or direction of the circuit wiring. If the wafer and circuit wiring defects are identified at the same time in the same machine vision recognition, it may not be possible to accurately distinguish and locate different types of defects, and it is impossible to comprehensively and accurately identify all types of defects on the surface of the circuit board. To this end, by performing machine vision recognition on the circuit board under different lighting conditions, the wafer defects and circuit wiring defects can be differentiated and directional recognized. Accordingly, the circuit board surface is photographed by machine vision in different regions under the illumination condition of a uniform surface beam with a single wavelength, and several first images of different regions of the circuit board surface are obtained, wherein the uniform surface beam can be, but not limited to, a beam of red or green wavelength, so that the uniform surface beam can perform uniform illumination on the corresponding region of the circuit board surface, so that the illuminated circuit board surface region can reflect the surface beam, and the reflected beam contains the surface morphology structure details of the wafer region corresponding to the circuit board surface; based on the shooting orientation information of the machine vision regional shooting, all the first images are spliced ​​to obtain the first circuit board image, thereby realizing the global wafer structure visual representation of the circuit board surface. Then, based on the light wavelength corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing matching the light wavelength to obtain the corresponding circuit board image, wherein the image layer extraction processing is performed by extracting the spectral data of the corresponding wavelength of the first circuit board image, which belongs to the conventional technical means in the field and is not described in detail here. The circuit board image is then processed with pixel contour recognition to obtain the first edge contour feature information of the alignment mark on the circuit board surface and the edge contour feature information of the morphological defect of the wafer part on the circuit board surface. This enables the alignment mark on the circuit board surface and the morphological defect on the wafer surface to be comprehensively and accurately identified.

[0115] In another embodiment, the circuit board is photographed by machine vision under a second lighting condition to obtain a second circuit board image; based on the lighting characteristic information of the second lighting condition, the second circuit board image is recognized to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board, including:

[0116] Performing machine vision scanning and photographing of the circuit board surface under white structured light illumination with a pattern of alternating light and dark stripes to obtain a second circuit board image corresponding to a global optical pattern formed by the circuit board surface after the white structured light is reflected;

[0117] After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board.

[0118] The beneficial effects of the above embodiments are that the circuit wiring defects on the surface of the circuit board mainly include wiring defects and solder joint defects, and the defect shapes they present are rich and varied. In order to comprehensively identify the circuit wiring defects, the surface of the circuit board is scanned and photographed by machine vision under the illumination conditions of white structured light with a pattern of alternating light and dark stripes, and a second circuit board image corresponding to the global optical pattern formed after the circuit board surface reflects the white structured light is obtained, that is, the circuit board surface is illuminated by structured light with a periodic pattern of alternating light and dark stripes formed by a white light source, so that the circuit board surface will adjust the light intensity of the structured light and reflect it to form a corresponding global optical pattern, so that the second circuit board image can comprehensively capture the global optical pattern. After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defects / solder point defects on the surface of the circuit board. The above-mentioned comparison processing is a common means of structured light detection and will not be described in detail here. The above-mentioned method can accurately locate and identify the wiring defects / solder point defects on the surface of the circuit board.

[0119] In another embodiment, based on the first alignment mark characteristic information and the second alignment mark characteristic information, the defect state information of the wafer portion and the wiring state characteristic information are compared in the same area to determine the relative relationship state between the wafer portion and the wiring portion of the circuit board; based on the relative relationship state, the defect bad point information on the surface of the circuit board is identified, and based on the defect bad point information, the quality state of the circuit board is determined, including:

[0120] Comparing the orientation of the first edge profile feature information and the second edge profile feature information to determine the orientation deviation between the two;

[0121] Based on the orientation deviation, the edge contour feature information of the morphology defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board are adjusted in terms of projection orientation on a two-dimensional plane, so as to project the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect to a corresponding two-dimensional plane space, thereby performing an orientation comparison of the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect in the same area, and obtaining a relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board;

[0122] Based on the relative distance state, the wafer defect-wiring defect / solder point defect pairs that affect the normal operation of the circuit board are identified from all wafer defects and all wiring defects / solder point defects of the circuit board, and are used as defective bad spots on the surface of the circuit board; then, based on the corresponding defective bad spot distribution density on the surface of the circuit board, it is determined whether the circuit board is a good circuit board or a defective circuit board.

[0123] The beneficial effect of the above embodiment is that the first edge contour feature information and the second edge contour feature information are both edge contour recognition results of the same alignment mark on the circuit board, but the placement orientation of the circuit board is not necessarily exactly the same when the machine vision shooting is performed under the first lighting condition and the second lighting condition, which makes the edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board have orientation differences. In order to facilitate the comparison of the edge contour feature information of the morphological defect of the wafer part and the edge contour feature information of the wiring defect / solder point defect in the same orientation space Analysis, first compare the orientation of the first edge profile feature information and the second edge profile feature information to determine the orientation deviation between the two, and then adjust the projection orientation of the morphological defect edge profile feature information of the wafer part on the surface of the circuit board and the wiring defect / solder point defect edge profile feature information on the surface of the circuit board on the two-dimensional plane based on the orientation deviation, so as to project the morphological defect edge profile feature information and the wiring defect / solder point defect edge profile feature information to the corresponding two-dimensional plane space, so as to obtain the relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board. Furthermore, the relative distance between the wafer defect and the wiring defect / solder point defect of the circuit board is compared with the preset distance threshold. If the relative distance is less than or equal to the preset distance threshold, the corresponding wafer defect and all wiring defects / solder point defects are determined as wafer defects-wiring defects / solder point defects that affect the normal operation of the circuit board, and used as defective bad spots on the surface of the circuit board. The defective bad pixel distribution density corresponding to the surface of the circuit board (i.e., the number of defective bad pixel distribution per unit area) is then compared with the preset density threshold. If the defective bad pixel distribution density is greater than the preset density threshold, the circuit board is determined to be a defective circuit board. Otherwise, the circuit board is determined to be a good circuit board, thereby quantitatively distinguishing the quality status of the circuit board.

[0124] In general, the circuit board defect detection system and method respectively perform machine vision photography and identification of the circuit board under the first lighting condition and the second lighting condition, obtain the first and second alignment mark feature information on the surface of the circuit board and the defect status information of the wafer part and the wiring status feature information, and can accurately differentiate and identify the defects existing in the wafer and the wiring through different lighting conditions, use the first and second alignment mark feature information to compare the defect status information of the wafer part and the wiring status feature information in the same area, associate and identify the defects existing in the wafer part and the wiring part of the circuit board, thereby identifying the defect bad point information that affects the normal working performance of the circuit board, providing a reliable basis for judging the quality status of the circuit board, and performing integrated and diversified defect detection on the circuit board by identifying the defects existing on the wafer and wiring solder joints of the circuit board, thereby realizing reliable performance quality screening of the circuit board.

[0125] The above is only a specific implementation of the present invention, and any other improvements made based on the concept of the present invention are considered to be within the protection scope of the present invention.

Claims

1. A circuit board defect detection system, characterized in that: include: A first visual shooting module, used for performing machine vision shooting of the circuit board under a first lighting condition to obtain a first circuit board image; A first image recognition module, configured to recognize the first circuit board image based on the lighting characteristic information of the first lighting condition, and obtain first alignment mark characteristic information on the surface of the circuit board and defect status information of a wafer portion; A second visual shooting module, used for performing machine vision shooting of the circuit board under a second lighting condition to obtain a second circuit board image; A second image recognition module, configured to recognize the second circuit board image based on the lighting characteristic information of the second lighting condition, and obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board; A wafer and wiring relative relationship determination module, used to compare the defect state information of the wafer part and the wiring state characteristic information in the same area based on the first alignment mark characteristic information and the second alignment mark characteristic information, and determine the relative relationship state between the wafer part and the wiring part of the circuit board; A circuit board quality determination module, used to identify defective bad pixel information on the surface of the circuit board based on the relative relationship state, and determine the quality state of the circuit board based on the defective bad pixel information; The first visual shooting module is used to perform machine vision shooting of the circuit board under the first lighting condition to obtain a first circuit board image, including: Under the condition of uniform surface light beam illumination with a single wavelength, the surface of the circuit board is photographed by machine vision in different regions to obtain a plurality of first images of different regions of the surface of the circuit board; based on the shooting orientation information of the machine vision regional photography, all the first images are spliced ​​to obtain a first circuit board image; The first image recognition module is used to recognize the first circuit board image based on the lighting characteristic information of the first lighting condition, and obtain the first alignment mark characteristic information on the surface of the circuit board and the defect state information of the wafer part, including: Based on the light wavelength corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing that matches the light wavelength to obtain a corresponding circuit board image; the circuit board image is subjected to pixel contour recognition processing to obtain first edge contour feature information of the alignment mark on the surface of the circuit board and edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board.

2. The circuit board defect detection system according to claim 1, characterized in that: The second visual shooting module is used to perform machine vision shooting of the circuit board under a second lighting condition to obtain a second circuit board image, including: Performing machine vision scanning and photographing of the circuit board surface under white structured light illumination with a pattern of alternating light and dark stripes to obtain a second circuit board image corresponding to a global optical pattern formed by the circuit board surface after the white structured light is reflected; The second image recognition module is used to recognize the second circuit board image based on the lighting characteristic information of the second lighting condition to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board, including: After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board.

3. The circuit board defect detection system according to claim 2, characterized in that: The wafer and wiring relative relationship determination module is used to compare the defect state information of the wafer part and the wiring state characteristic information in the same area based on the first alignment mark characteristic information and the second alignment mark characteristic information, and determine the relative relationship state between the wafer part and the wiring part of the circuit board, including: Comparing the first edge profile feature information and the second edge profile feature information, the orientation of the comparison mark is compared to determine the orientation deviation between the two; Based on the orientation deviation, the edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board are adjusted in terms of projection orientation on a two-dimensional plane, so as to project the edge contour feature information of the morphological defect and the edge contour feature information of the wiring defect / solder point defect to a corresponding two-dimensional plane space, thereby performing a same-region orientation comparison between the edge contour feature information of the morphological defect and the edge contour feature information of the wiring defect / solder point defect, and obtaining a relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board; the circuit board quality determination module is used to identify the existence of defective bad point information on the surface of the circuit board based on the relative relationship state, and determine the quality state of the circuit board based on the defective bad point information, including: Based on the relative distance state, the wafer defect-wiring defect / solder point defect pairs that affect the normal operation of the circuit board are identified from all wafer defects and all wiring defects / solder point defects of the circuit board, and are used as defective bad spots on the surface of the circuit board; and based on the distribution density of defective bad spots corresponding to the surface of the circuit board, it is determined whether the circuit board is a good circuit board or a defective circuit board.

4. The circuit board defect detection system according to claim 3, characterized in that: It also includes a shooting positioning module, which is used to position the shooting of the first visual shooting module and the second visual shooting module, so as to ensure that the shooting positions of the first visual shooting module and the second visual shooting module are consistent, and includes: Step S1: There are four vertex positioning points on the shooting table where the circuit board is placed, and each positioning point can be recognized by the first visual shooting module and the second visual shooting module. The following formula (1) is used to determine whether the captured circuit board image has image distortion according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the second visual shooting module. J=F{[(X1-X2,Y1-Y2)∥(X3-X4,Y3-Y4)]&&[(X1-X3,Y1-Y3)∥(X2-X4,Y2-Y4)]}(1) In the above formula (1), J represents the judgment value for judging whether the captured circuit board image has image distortion; (X1, Y1) represents the coordinates of the upper left corner positioning point in the circuit board image captured by the camera module; (X2, Y2) represents the coordinates of the upper right corner positioning point in the circuit board image captured by the camera module; (X3, Y3) represents the coordinates of the lower left corner positioning point in the circuit board image captured by the camera module; (X4, Y4) represents the coordinates of the lower right corner positioning point in the circuit board image captured by the camera module; ∥ represents parallel to; && represents the logical relationship and; F{} represents the judgment function, if the formula in the brackets is established, the function value is 1, if the formula in the brackets is not established, the function value is 0; If J = 1, it means that there is no image distortion in the captured circuit board image; If J = 0, it means that the captured circuit board image has image distortion; If there is image distortion, the shooting module is adjusted and the above step S1 is repeated until the circuit board images shot by the first visual shooting module and the second visual shooting module have no image distortion; Step S2: when there is no image distortion in the circuit board images captured by the first visual shooting module and the second visual shooting module, the shooting parameters of the first visual shooting module and the second visual shooting module are set to be consistent, and then the following formula (2) is used to determine whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module. In the above formula (2), P represents a determination value for determining whether the shooting positions of the first visual shooting module and the second visual shooting module are consistent; [x1(a), y1(a)] represents the coordinates of the a-th vertex positioning point in the circuit board image captured by the first visual shooting module; [x2(a), y2(a)] represents the coordinates of the a-th vertex positioning point in the circuit board image captured by the second visual shooting module; If P=1, it means that the shooting positions of the first visual shooting module and the second visual shooting module are consistent; If P=0, it means that the shooting positions of the first visual shooting module and the second visual shooting module are inconsistent; Step S3: When the shooting positions of the first visual shooting module and the second visual shooting module are inconsistent, the following formula (3) is used to control the direction of the up and down movement adjustment of the second visual shooting module according to the coordinate positions of the four vertex positioning points in the circuit board image captured by the first visual shooting module and the coordinate positions of the four vertex positioning points in the circuit board image captured by the second visual shooting module. In the above formula (3), R represents the direction control value of the up and down movement adjustment of the second visual shooting module; L{,} represents the distance value between the two coordinate points in the brackets; If R=1, it means that the second visual shooting module is controlled to move upward and the above steps S2 and S3 are repeated until the shooting positions of the first visual shooting module and the second visual shooting module are consistent; If R=0, it means that the second visual shooting module is controlled to move downward and the above steps S2 and S3 are repeated until the shooting positions of the first visual shooting module and the second visual shooting module are consistent.

5. A circuit board defect detection method, characterized in that: include: Performing machine vision photography on the circuit board under a first lighting condition to obtain a first circuit board image; Based on the lighting characteristic information of the first lighting condition, the first circuit board image is recognized to obtain first alignment mark characteristic information on the surface of the circuit board and defect status information of a wafer portion; Performing machine vision photography on the circuit board under a second lighting condition to obtain a second circuit board image; recognizing the second circuit board image based on lighting characteristic information of the second lighting condition to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board; Based on the first alignment mark characteristic information and the second alignment mark characteristic information, the defect state information of the wafer part and the wiring state characteristic information are compared in the same area to determine the relative relationship between the wafer part and the wiring part of the circuit board; Based on the relative relationship state, identifying defective bad pixel information on the surface of the circuit board, and determining the quality state of the circuit board based on the defective bad pixel information; Performing machine vision photography on a circuit board under a first lighting condition to obtain a first circuit board image; recognizing the first circuit board image based on lighting feature information of the first lighting condition to obtain first alignment mark feature information on the surface of the circuit board and defect status information of a wafer portion, including: Under the condition of uniform surface light beam illumination with a single wavelength, the surface of the circuit board is photographed by machine vision in different regions to obtain a plurality of first images of different regions of the surface of the circuit board; based on the shooting orientation information of the machine vision regional photography, all the first images are spliced ​​to obtain a first circuit board image; Based on the light wavelength corresponding to the uniform surface beam illumination condition, the first circuit board image is subjected to image layer extraction processing that matches the light wavelength to obtain a corresponding circuit board image; the circuit board image is subjected to pixel contour recognition processing to obtain first edge contour feature information of the alignment mark on the surface of the circuit board and edge contour feature information of the morphological defect of the wafer part on the surface of the circuit board.

6. The circuit board defect detection method according to claim 5, characterized in that: Performing machine vision photography on the circuit board under a second lighting condition to obtain a second circuit board image; recognizing the second circuit board image based on lighting characteristic information of the second lighting condition to obtain second alignment mark characteristic information and wiring state characteristic information on the surface of the circuit board, including: Performing machine vision scanning and photographing of the circuit board surface under white structured light illumination with a pattern of alternating light and dark stripes to obtain a second circuit board image corresponding to a global optical pattern formed by the circuit board surface after the white structured light is reflected; After edge pixel enhancement processing is performed on the second circuit board image, the optical pattern texture size distribution information and light intensity distribution information of the second circuit board image are obtained; the size distribution information and light intensity distribution information of the light and dark stripes corresponding to the white structured light illumination condition are compared with the optical pattern texture size and light intensity distribution information of the second circuit board image, respectively, to obtain the second edge contour feature information of the alignment mark on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board.

7. The circuit board defect detection method according to claim 6, characterized in that: Based on the first alignment mark characteristic information and the second alignment mark characteristic information, the defect state information of the wafer part and the wiring state characteristic information are compared in the same area to determine the relative relationship between the wafer part and the wiring part of the circuit board; Based on the relative relationship state, identifying defective bad pixel information on the surface of the circuit board, and based on the defective bad pixel information, determining the quality state of the circuit board, including: Comparing the first edge profile feature information and the second edge profile feature information, the orientation of the comparison mark is compared to determine the orientation deviation between the two; Based on the orientation deviation, the edge contour feature information of the morphology defect of the wafer part on the surface of the circuit board and the edge contour feature information of the wiring defect / solder point defect on the surface of the circuit board are adjusted with respect to the projection orientation in the two-dimensional plane, so as to project the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect into the corresponding two-dimensional plane space, so as to compare the edge contour feature information of the morphology defect and the edge contour feature information of the wiring defect / solder point defect in the same area, and obtain the relative distance state between the wafer defect and the wiring defect / solder point defect of the circuit board; based on the relative distance state, the wafer defect-wiring defect / solder point defect pair that affects the normal operation of the circuit board is identified from all the wafer defects and all the wiring defects / solder point defects of the circuit board, and this is used as the defective bad point existing on the surface of the circuit board; and based on the distribution density of the defective bad points corresponding to the surface of the circuit board, it is determined whether the circuit board is a good circuit board or a defective circuit board.

Images