Mainboard defect detection method, device and equipment based on three-dimensional imaging, medium and program product

Abstract

The invention relates to the technical field of microscopic imaging processing, in particular to a mainboard defect detection method and device based on three-dimensional imaging, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring imaging data of a target object; processing the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data; registering the three-dimensional point cloud data, and performing partition mask processing to obtain a plurality of sub-regions; based on the mapping relationship between the three-dimensional point cloud data and the reference drawing, determining the region type of the sub-region, preprocessing the three-dimensional point cloud data based on a preprocessing algorithm corresponding to the region type to obtain application point cloud data, the preprocessing algorithm being matched with the associated defect type in the sub-region; and reconstructing a three-dimensional reconstruction model of the target object based on application of the point cloud data, and performing defect detection on the target object to obtain defect feature information. By adopting the method, the adaptability and the detection precision of defect detection can be improved.

Description

Technical Field

[0001] This application relates to the field of microscopic imaging processing technology, and in particular to a method, apparatus, computer equipment, storage medium and computer program product for motherboard defect detection based on three-dimensional imaging. Background Technology

[0002] In the semiconductor manufacturing industry, the motherboard (such as a printed circuit board (PCB) or integrated circuit substrate) serves as the core carrier of electronic devices, and its quality directly affects the performance and reliability of the final product. However, in complex manufacturing processes, motherboards may develop various defects due to material defects, processing errors, or environmental factors, such as open circuits, short circuits, poor solder joints, broken circuits, foreign matter contamination, or interlayer alignment deviations. If these defects are not detected in time, they will lead to abnormal equipment function, decreased yield, and even safety hazards, causing huge economic losses. Therefore, high-precision and high-efficiency defect detection technology has become a key requirement in the semiconductor industry. The core objective of defect detection is to identify and locate defects as early as possible during the manufacturing process so that they can be repaired or defective products can be discarded, thereby improving product yield and reducing production costs. As semiconductor devices develop towards higher integration and miniaturization, the linewidth and spacing of motherboards continue to shrink (e.g., into the micrometer or even nanometer scale), making it difficult for traditional manual visual inspection or simple optical inspection to meet the accuracy requirements. In addition, the stringent reliability standards of modern electronic devices (such as automotive electronics and aerospace) have further driven the development of automated and intelligent defect detection technologies.

[0003] Among related technologies, mainstream defect detection techniques are mainly based on optical imaging, electrical testing, or a combination of both. Optical inspection acquires images of the motherboard surface using high-resolution cameras or scanning devices, and then uses image processing algorithms (such as edge detection, template matching, deep learning, etc.) to compare the standard image with the image under test to identify abnormal areas. Electrical testing uses probes or flying probes to detect electrical parameters such as the continuity and impedance of circuits to determine whether there are problems such as open circuits or short circuits. In addition, X-ray inspection, infrared thermal imaging, and other technologies are also used to detect internal structural defects or abnormal heat distribution.

[0004] However, current motherboard defect detection methods have the following technical problems: Existing defect detection methods rely on image analysis algorithms for processing. However, commonly used image analysis algorithms have the problem of falsely filtering out defect features during processing, causing defects to be identified as noise points and ignored, resulting in a low defect detection rate. Summary of the Invention

[0005] Therefore, it is necessary to provide a motherboard defect detection method, device, computer equipment, computer-readable storage medium, and computer program product based on three-dimensional imaging that can improve the adaptability and detection accuracy of defect detection, in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a motherboard defect detection method based on three-dimensional imaging. The method includes: Acquire imaging data of the target object to be processed, including visible light detection images, structured light detection images, and transmission detection images; The imaging data is processed based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; The 3D point cloud data is registered based on a reference drawing that matches the target object, and the 3D point cloud data is partitioned and masked to obtain several sub-regions. Based on the mapping relationship between the 3D point cloud data and the reference drawing, the region type of the sub-region is determined. The 3D point cloud data in the sub-region is preprocessed based on the preprocessing algorithm corresponding to the region type to obtain the application point cloud data. The preprocessing algorithm is matched with the defect type associated in the sub-region. Based on the application point cloud data, a three-dimensional reconstruction model of the target object is reconstructed, and defect detection is performed on the target object according to the three-dimensional reconstruction model to obtain defect feature information.

[0007] In one embodiment, the registration of the 3D point cloud data based on a reference drawing matching the target object, and the partitioning and masking of the 3D point cloud data to obtain several sub-regions, include: The reference image is subjected to three-dimensional meshing to obtain mesh cells corresponding to different functional layers in the reference drawing. The mask group number is initialized based on the grid cell, and the three-dimensional point cloud data is traversed based on the spatial index relationship between the three-dimensional point cloud data and the reference drawing, and each point cloud is divided into the corresponding target grid cell.

[0008] In one embodiment, the preprocessing of the 3D point cloud data within the sub-region based on a preprocessing algorithm corresponding to the region type to obtain application point cloud data includes: Based on the region type, the unique defect types existing in the sub-region are determined, and based on the unique defect types and the common defect types corresponding to the target object, a defect feature library corresponding to each sub-region is constructed. Based on the defect feature library, the three-dimensional point cloud data is traversed, and target point cloud data whose matching degree with the defect structure in the defect feature library exceeds a preset threshold are marked as protection points. The protection points are then isolated in the preprocessing.

[0009] In one embodiment, the preprocessing of the 3D point cloud data within the sub-region based on a preprocessing algorithm corresponding to the region type to obtain application point cloud data includes: Based on the region type, determine the baseline denoising threshold for different sub-regions; Different adjustment values ​​are set for different PCB areas within the sub-region, and the baseline denoising threshold is adjusted based on the adjustment values ​​to determine the strictness of the partitioned filtering within the sub-region.

[0010] In one embodiment, setting different adjustment values ​​for different PCB areas within the sub-region includes: The sub-region is divided into several PCB regions based on preset partitioning constraints. The partitioning constraints include geometric height partitioning, surface normal partitioning, material type partitioning, and functional area partitioning. The adjustment value is set based on the noise risk and defect risk of the PCB area. The higher the noise risk of the PCB area, the lower the adjustment value, and the higher the defect risk of the PCB area, the higher the adjustment value.

[0011] In one embodiment, after reconstructing a 3D reconstruction model of the target object based on the application point cloud data, and performing defect detection on the target object according to the 3D reconstruction model to obtain defect feature information, the method further includes: Obtain the actual defect marking information and defect feature information of the target object; When the difference rate between the defect labeling information and the defect feature information exceeds a preset threshold, the benchmark denoising threshold is adjusted. Secondly, this application also provides a motherboard defect detection device based on three-dimensional imaging. The device includes: An imaging data module is used to acquire imaging data of the target object to be processed, the imaging data including visible light detection images, structured light detection images and transmission detection images; The point cloud data module is used to process the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; The registration and partitioning module is used to register the 3D point cloud data based on the reference drawing that matches the target object, and to perform partitioning masking on the 3D point cloud data to obtain several sub-regions; The partition preprocessing module is used to determine the region type of the sub-region based on the mapping relationship between the 3D point cloud data and the reference drawing, and to preprocess the 3D point cloud data in the sub-region based on the preprocessing algorithm corresponding to the region type to obtain application point cloud data. The preprocessing algorithm is matched with the defect type associated in the sub-region. The 3D reconstruction module is used to reconstruct a 3D reconstruction model of the target object based on the application point cloud data, and to perform defect detection on the target object according to the 3D reconstruction model to obtain defect feature information.

[0012] In one embodiment, the registration partitioning module includes: The grid unit module is used to perform three-dimensional gridding processing on the reference image to obtain grid units corresponding to different functional layers in the reference drawing. The point cloud partitioning module is used to initialize the number of mask groups based on the grid cells, traverse the three-dimensional point cloud data based on the spatial index relationship between the three-dimensional point cloud data and the reference drawing, and partition each point cloud into the corresponding target grid cell.

[0013] In one embodiment, the partition preprocessing module includes: The defect feature library module is used to determine the unique defect types existing in the sub-region according to the region type, and to construct a defect feature library corresponding to each sub-region based on the unique defect types and the common defect types corresponding to the target object. The protection point module is used to traverse the three-dimensional point cloud data based on the defect feature library, mark the target point cloud data with a matching degree of more than a preset threshold with the defect structure in the defect feature library as protection points, and isolate the protection points in the preprocessing.

[0014] In one embodiment, the partition preprocessing module includes: A baseline threshold module is used to determine a baseline denoising threshold for different sub-regions based on the region type. The adjustment value module is used to set different adjustment values ​​for different PCB areas within the sub-region, and adjust the benchmark denoising threshold based on the adjustment values ​​to determine the strictness of the partitioned filtering within the sub-region.

[0015] In one embodiment, the adjustment value module includes: The secondary partitioning module is used to perform secondary partitioning of the sub-region based on preset partitioning constraints to obtain several PCB regions. The partitioning constraints include geometric height partitioning, surface normal partitioning, material type partitioning, and functional area partitioning. The adjustment value matching module is used to set matching adjustment values ​​based on the noise risk and defect risk of the PCB area. The higher the noise risk of the PCB area, the lower the adjustment value, and the higher the defect risk of the PCB area, the higher the adjustment value.

[0016] In one embodiment, after the three-dimensional reconstruction, the method further includes: The defect marking module is used to obtain the actual defect marking information and defect feature information of the target object; The feedback adjustment module is used to adjust the benchmark denoising threshold when the difference rate between the defect marking information and the defect feature information exceeds a preset threshold.

[0017] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of a motherboard defect detection method based on three-dimensional imaging as described in any embodiment of the first aspect.

[0018] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of a motherboard defect detection method based on three-dimensional imaging as described in any embodiment of the first aspect.

[0019] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of a motherboard defect detection method based on three-dimensional imaging as described in any embodiment of the first aspect.

[0020] The aforementioned motherboard defect detection method, apparatus, computer equipment, storage medium, and computer program product based on three-dimensional imaging, derived from the technical features in the claims, can achieve the following beneficial effects corresponding to the technical problems in the background art: This application provides a motherboard defect detection method based on three-dimensional imaging, comprising: acquiring imaging data of a target object to be processed, the imaging data including visible light detection images, structured light detection images, and transmission detection images; processing the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; registering the three-dimensional point cloud data based on a reference drawing matching the target object, and performing partitioning masking on the three-dimensional point cloud data to obtain several sub-regions; determining the region type of the sub-regions based on the mapping relationship between the three-dimensional point cloud data and the reference drawing; preprocessing the three-dimensional point cloud data in the sub-regions based on a preprocessing algorithm corresponding to the region type to obtain application point cloud data, wherein the preprocessing algorithm matches the defect type associated with the sub-regions; reconstructing a three-dimensional reconstruction model of the target object based on the application point cloud data; and performing defect detection on the target object based on the three-dimensional reconstruction model to obtain defect feature information. In implementation, during image fusion processing, multimodal image fusion allows for the complementarity of various imaging data. Visible light detection preserves the color features of the target object's surface, structured light provides high-precision geometric information, and transmission images reveal the target object's internal structural features. This facilitates the extraction of multidimensional features associated with defects, preventing defect features from being masked by single-modal noise and reducing the possibility of subsequent over-filtering. Subsequently, partitioning masks and adaptive preprocessing for different regions avoid the possibility of misfiltering local defect points due to using uniform denoising parameters globally. This helps match different denoising thresholds according to the defect type in different regions, preserving defect features while maintaining denoising performance, thus improving the defect retention rate. Finally, defect feature detection based on a 3D reconstruction model facilitates accurate identification of defect features at the 3D model level, improving the efficiency and stability of defect detection. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the first process of a motherboard defect detection method based on three-dimensional imaging in one embodiment; Figure 2 This is a schematic diagram of the second process of a motherboard defect detection method based on three-dimensional imaging in another embodiment; Figure 3 This is a schematic diagram of the third process of a motherboard defect detection method based on three-dimensional imaging in another embodiment; Figure 4 This is a schematic diagram of the fourth process of a motherboard defect detection method based on three-dimensional imaging in another embodiment; Figure 5 This is a schematic diagram of the fifth process of a motherboard defect detection method based on three-dimensional imaging in another embodiment; Figure 6 This is a schematic diagram of the sixth process of a motherboard defect detection method based on three-dimensional imaging in another embodiment; Figure 7 This is a structural block diagram of a motherboard defect detection device based on three-dimensional imaging in one embodiment; Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0024] Among related technologies, mainstream defect detection techniques are mainly based on optical imaging, electrical testing, or a combination of both. Optical inspection acquires images of the motherboard surface using high-resolution cameras or scanning devices, and then uses image processing algorithms (such as edge detection, template matching, deep learning, etc.) to compare the standard image with the image under test to identify abnormal areas. Electrical testing uses probes or flying probes to detect electrical parameters such as the continuity and impedance of circuits to determine whether there are problems such as open circuits or short circuits. In addition, X-ray inspection, infrared thermal imaging, and other technologies are also used to detect internal structural defects or abnormal heat distribution.

[0025] However, current motherboard defect detection methods have the following technical problems: Existing defect detection methods rely on image analysis algorithms for processing. However, commonly used image analysis algorithms have the problem of falsely filtering out defect features during processing, causing defects to be identified as noise points and ignored, resulting in a low defect detection rate.

[0026] In one embodiment, such as Figure 1 As shown, a motherboard defect detection method based on three-dimensional imaging is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps: Step 102: Obtain imaging data of the target object to be processed, including visible light detection images, structured light detection images, and transmission detection images.

[0027] Step 104: Process the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object.

[0028] Step 106: Register the 3D point cloud data based on the reference drawing that matches the target object, and perform partitioning masking on the 3D point cloud data to obtain several sub-regions.

[0029] Step 108: Based on the mapping relationship between the 3D point cloud data and the reference drawing, determine the region type of the sub-region, and preprocess the 3D point cloud data in the sub-region based on the preprocessing algorithm corresponding to the region type to obtain application point cloud data. The preprocessing algorithm is matched with the defect type associated in the sub-region.

[0030] Step 1010: Reconstruct a three-dimensional reconstruction model of the target object based on the application point cloud data, and perform defect detection on the target object according to the three-dimensional reconstruction model to obtain defect feature information.

[0031] In the aforementioned motherboard defect detection method based on three-dimensional imaging, reasonable derivation based on the technical features in the embodiments achieves the beneficial effect of solving the technical problems raised in the background art: This application provides a motherboard defect detection method based on three-dimensional imaging, comprising: acquiring imaging data of a target object to be processed, the imaging data including visible light detection images, structured light detection images, and transmission detection images; processing the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; registering the three-dimensional point cloud data based on a reference drawing matching the target object, and performing partitioning masking on the three-dimensional point cloud data to obtain several sub-regions; determining the region type of the sub-regions based on the mapping relationship between the three-dimensional point cloud data and the reference drawing; preprocessing the three-dimensional point cloud data in the sub-regions based on a preprocessing algorithm corresponding to the region type to obtain application point cloud data, wherein the preprocessing algorithm matches the defect type associated with the sub-regions; reconstructing a three-dimensional reconstruction model of the target object based on the application point cloud data; and performing defect detection on the target object based on the three-dimensional reconstruction model to obtain defect feature information. In implementation, during image fusion processing, multimodal image fusion allows for the complementarity of various imaging data. Visible light detection preserves the color features of the target object's surface, structured light provides high-precision geometric information, and transmission images reveal the target object's internal structural features. This facilitates the extraction of multidimensional features associated with defects, preventing defect features from being masked by single-modal noise and reducing the possibility of subsequent over-filtering. Subsequently, partitioning masks and adaptive preprocessing for different regions avoid the possibility of misfiltering local defect points due to using uniform denoising parameters globally. This helps match different denoising thresholds according to the defect type in different regions, preserving defect features while maintaining denoising performance, thus improving the defect retention rate. Finally, defect feature detection based on a 3D reconstruction model facilitates accurate identification of defect features at the 3D model level, improving the efficiency and stability of defect detection.

[0032] In one embodiment, such as Figure 2 As shown, step 106 includes: Step 202: Perform three-dimensional meshing processing on the reference image to obtain mesh cells corresponding to different functional layers in the reference drawing; Step 204: Initialize the number of mask groups based on the grid cells, traverse the three-dimensional point cloud data based on the spatial index relationship between the three-dimensional point cloud data and the reference drawing, and divide each point cloud into the corresponding target grid cells.

[0033] In this embodiment, the precise partitioning of point cloud data is achieved through the three-dimensional meshing and spatial index mapping of the reference drawing. This helps to physically divide various functional layers, such as copper foil, solder mask, and substrate, based on CAD drawings. At the same time, the partitioning based on the drawing avoids defects crossing regions, thus preventing discrimination errors caused by missing defect structures. This ensures that defects fall into the same mesh, ultimately helping to improve the accuracy of defect detection.

[0034] In one embodiment, such as Figure 3 As shown, step 108 includes: Step 302: Determine the unique defect types existing in the sub-regions according to the region type, and construct a defect feature library corresponding to each sub-region based on the unique defect types and the common defect types corresponding to the target object.

[0035] Step 304: Based on the defect feature library, traverse the three-dimensional point cloud data, mark the target point cloud data with a matching degree of more than a preset threshold with the defect structure in the defect feature library as protection points, and isolate the protection points in the preprocessing.

[0036] In this embodiment, the defect feature library constructed through prior knowledge helps to match point cloud data with the defect feature library, thereby identifying target point cloud data with high matching degree and marking them as protection points. This allows protection points to be isolated in the denoising process, preventing point clouds related to defects from being filtered out as noise, and helps to improve the stability of defect detection.

[0037] In one embodiment, such as Figure 4 As shown, step 108 includes: Step 402: Determine the baseline denoising threshold for different sub-regions based on the region type.

[0038] Step 404: Set different adjustment values ​​for different PCB areas within the sub-region, adjust the benchmark denoising threshold based on the adjustment values, and determine the strictness of the partitioned filtering within the sub-region.

[0039] In this embodiment, since differences in materials, component density, and minute defects can lead to differences in noise characteristics, different reference noise thresholds can be set for different regions by region type. Subsequently, different differential adjustments can be made to the local point cloud, which helps to adjust the noise threshold in high-defect regions and improve the accuracy of defect detection.

[0040] In one embodiment, such as Figure 5 As shown, step 304 includes: Step 502: Perform secondary partitioning on the sub-region based on preset partitioning constraints to obtain several PCB regions. The partitioning constraints include geometric height partitioning, surface normal partitioning, material type partitioning, and functional area partitioning.

[0041] Step 504: Set a matching adjustment value based on the noise risk and defect risk of the PCB area. The higher the noise risk of the PCB area, the lower the adjustment value. The higher the defect risk of the PCB area, the higher the adjustment value.

[0042] In this embodiment, adjusting the denoising threshold from the perspectives of defect risk and noise risk helps to reduce the possibility of defect points being filtered out while preserving the denoising effect, thereby improving the stability and accuracy of defect detection.

[0043] In one embodiment, such as Figure 6 As shown, after step 1010, the method further includes: Step 602: Obtain the actual defect marking information and defect feature information of the target object.

[0044] Step 604: When the difference rate between the defect marking information and the defect feature information exceeds a preset threshold, adjust the benchmark denoising threshold.

[0045] In this embodiment, the denoising threshold is adjusted by feedback using actual defect marking information. This helps to dynamically adjust the denoising threshold based on actual defect detection, dynamically balancing the past noise and under-denoising conditions, and improving the flexibility of defect detection.

[0046] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0047] Based on the same inventive concept, this application also provides a motherboard defect detection device based on three-dimensional imaging for implementing the aforementioned motherboard defect detection method based on three-dimensional imaging. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of a motherboard defect detection device based on three-dimensional imaging provided below can be found in the above-described limitations of the motherboard defect detection method based on three-dimensional imaging, and will not be repeated here.

[0048] In one embodiment, such as Figure 7 As shown, a motherboard defect detection device based on three-dimensional imaging is provided, including: an imaging data module, a point cloud data module, a registration and partitioning module, a partitioning preprocessing module, and a three-dimensional reconstruction module, wherein: An imaging data module is used to acquire imaging data of the target object to be processed, the imaging data including visible light detection images, structured light detection images and transmission detection images; The point cloud data module is used to process the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; The registration and partitioning module is used to register the 3D point cloud data based on the reference drawing that matches the target object, and to perform partitioning masking on the 3D point cloud data to obtain several sub-regions; The partition preprocessing module is used to determine the region type of the sub-region based on the mapping relationship between the 3D point cloud data and the reference drawing, and to preprocess the 3D point cloud data in the sub-region based on the preprocessing algorithm corresponding to the region type to obtain application point cloud data. The preprocessing algorithm is matched with the defect type associated in the sub-region. The 3D reconstruction module is used to reconstruct a 3D reconstruction model of the target object based on the application point cloud data, and to perform defect detection on the target object according to the 3D reconstruction model to obtain defect feature information.

[0049] In one embodiment, the registration partitioning module includes: The grid unit module is used to perform three-dimensional gridding processing on the reference image to obtain grid units corresponding to different functional layers in the reference drawing. The point cloud partitioning module is used to initialize the number of mask groups based on the grid cells, traverse the three-dimensional point cloud data based on the spatial index relationship between the three-dimensional point cloud data and the reference drawing, and partition each point cloud into the corresponding target grid cell.

[0050] In one embodiment, the partition preprocessing module includes: The defect feature library module is used to determine the unique defect types existing in the sub-region according to the region type, and to construct a defect feature library corresponding to each sub-region based on the unique defect types and the common defect types corresponding to the target object. The protection point module is used to traverse the three-dimensional point cloud data based on the defect feature library, mark the target point cloud data with a matching degree of more than a preset threshold with the defect structure in the defect feature library as protection points, and isolate the protection points in the preprocessing.

[0051] In one embodiment, the partition preprocessing module includes: A baseline threshold module is used to determine a baseline denoising threshold for different sub-regions based on the region type. The adjustment value module is used to set different adjustment values ​​for different PCB areas within the sub-region, and adjust the benchmark denoising threshold based on the adjustment values ​​to determine the strictness of the partitioned filtering within the sub-region.

[0052] In one embodiment, the adjustment value module includes: The secondary partitioning module is used to perform secondary partitioning of the sub-region based on preset partitioning constraints to obtain several PCB regions. The partitioning constraints include geometric height partitioning, surface normal partitioning, material type partitioning, and functional area partitioning. The adjustment value matching module is used to set matching adjustment values ​​based on the noise risk and defect risk of the PCB area. The higher the noise risk of the PCB area, the lower the adjustment value, and the higher the defect risk of the PCB area, the higher the adjustment value.

[0053] In one embodiment, after the three-dimensional reconstruction, the method further includes: The defect marking module is used to obtain the actual defect marking information and defect feature information of the target object; The feedback adjustment module is used to adjust the benchmark denoising threshold when the difference rate between the defect marking information and the defect feature information exceeds a preset threshold.

[0054] The modules in the aforementioned motherboard defect detection device based on three-dimensional imaging can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0055] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a motherboard defect detection method based on three-dimensional imaging. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0056] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0057] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0058] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0059] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0060] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0061] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0062] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A motherboard defect detection method based on three-dimensional imaging, characterized in that, The method includes: Acquire imaging data of the target object to be processed, including visible light detection images, structured light detection images, and transmission detection images; The imaging data is processed based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; The 3D point cloud data is registered based on a reference drawing that matches the target object, and the 3D point cloud data is partitioned and masked to obtain several sub-regions. Based on the mapping relationship between the 3D point cloud data and the reference drawing, the region type of the sub-region is determined. The 3D point cloud data in the sub-region is preprocessed based on the preprocessing algorithm corresponding to the region type to obtain the application point cloud data. The preprocessing algorithm is matched with the defect type associated in the sub-region. Based on the application point cloud data, a three-dimensional reconstruction model of the target object is reconstructed, and defect detection is performed on the target object according to the three-dimensional reconstruction model to obtain defect feature information.

2. The method according to claim 1, characterized in that, The 3D point cloud data is registered based on a reference drawing that matches the target object, and the 3D point cloud data is then partitioned and masked to obtain several sub-regions, including: The reference image is subjected to three-dimensional meshing to obtain mesh cells corresponding to different functional layers in the reference drawing. The mask group number is initialized based on the grid cell, and the three-dimensional point cloud data is traversed based on the spatial index relationship between the three-dimensional point cloud data and the reference drawing, and each point cloud is divided into the corresponding target grid cell.

3. The method according to claim 1, characterized in that, The preprocessing algorithm based on the region type is used to preprocess the 3D point cloud data within the sub-region to obtain application point cloud data, including: Based on the region type, the unique defect types existing in the sub-region are determined, and based on the unique defect types and the common defect types corresponding to the target object, a defect feature library corresponding to each sub-region is constructed. Based on the defect feature library, the three-dimensional point cloud data is traversed, and target point cloud data whose matching degree with the defect structure in the defect feature library exceeds a preset threshold are marked as protection points. The protection points are then isolated in the preprocessing.

4. The method according to claim 1, characterized in that, The preprocessing algorithm based on the region type is used to preprocess the 3D point cloud data within the sub-region to obtain application point cloud data, including: Based on the region type, determine the baseline denoising threshold for different sub-regions; Different adjustment values ​​are set for different PCB areas within the sub-region, and the baseline denoising threshold is adjusted based on the adjustment values ​​to determine the strictness of the partitioned filtering within the sub-region.

5. The method according to claim 4, characterized in that, Setting different adjustment values ​​for different PCB areas within the sub-region includes: The sub-region is divided into several PCB regions based on preset partitioning constraints. The partitioning constraints include geometric height partitioning, surface normal partitioning, material type partitioning, and functional area partitioning. The adjustment value is set based on the noise risk and defect risk of the PCB area. The higher the noise risk of the PCB area, the lower the adjustment value, and the higher the defect risk of the PCB area, the higher the adjustment value.

6. The method according to claim 4, characterized in that, After reconstructing a 3D reconstruction model of the target object based on the application point cloud data, and performing defect detection on the target object according to the 3D reconstruction model to obtain defect feature information, the method further includes: Obtain the actual defect marking information and defect feature information of the target object; When the difference rate between the defect labeling information and the defect feature information exceeds a preset threshold, the benchmark denoising threshold is adjusted.

7. A motherboard defect detection device based on three-dimensional imaging, characterized in that, The device includes: An imaging data module is used to acquire imaging data of the target object to be processed, the imaging data including visible light detection images, structured light detection images and transmission detection images; The point cloud data module is used to process the imaging data based on a preset image fusion algorithm to obtain three-dimensional point cloud data corresponding to the target object; The registration and partitioning module is used to register the 3D point cloud data based on the reference drawing that matches the target object, and to perform partitioning masking on the 3D point cloud data to obtain several sub-regions; The partition preprocessing module is used to determine the region type of the sub-region based on the mapping relationship between the 3D point cloud data and the reference drawing, and to preprocess the 3D point cloud data in the sub-region based on the preprocessing algorithm corresponding to the region type to obtain application point cloud data. The preprocessing algorithm is matched with the defect type associated in the sub-region. The 3D reconstruction module is used to reconstruct a 3D reconstruction model of the target object based on the application point cloud data, and to perform defect detection on the target object according to the 3D reconstruction model to obtain defect feature information.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images