Circuit board defect detection method, device and equipment and storage medium

Abstract

The invention provides a circuit board defect detection method and device, equipment and a storage medium, and the method comprises the steps: obtaining an image of a bonding pad region of a to-be-detected circuit board, recognizing the actual contour of a bonding pad in the image, and generating a virtual contour corresponding to the actual contour; generating a visual detection result of the bonding pad based on the actual contour, the virtual contour and a preset visual detection caliper gauge; and fusing the visual detection result and the AI detection result of the bonding pad to obtain a complete detection result of the bonding pad. According to the technical scheme provided by the invention, the defects, such as tiny size deviation and tiny deformation, possibly caused by unclean development, of the bonding pad can be found more comprehensively and meticulously, the detection precision is greatly improved, the occurrence of missing detection and false detection conditions is effectively avoided, and the detection efficiency of the circuit board defects is improved. The problems existing in the bonding pad area can be found in time, so that the problems existing in the developing link are analyzed, and the production quality of the circuit board is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of defect detection, and in particular to a circuit board defect detection method, device and storage medium. Background Art

[0002] Printed Circuit Board (PCB), also known as printed circuit board, is a basic component that uses printing technology to form conductive circuits and component connection points on an insulating material substrate, used to install and connect various electronic components, thereby realizing electrical connections and functions in electronic devices. In the manufacturing process of PCB, development is a key process, which is to remove the unexposed photoresist from the exposed circuit board through a developer, so that the circuit pattern that needs to be etched or electroplated can be clearly displayed. When there is residual photoresist that should not exist on the circuit board after development or there are traces of photoresist that have not been completely removed, it is called unclean development.

[0003] The inventors found that when using artificial intelligence (AI) models to identify defects on PCBs, the defect features caused by unclean development are not obvious. For example, the slight convex or concave deformation on the gold surface cannot be effectively detected, resulting in a high miss rate of defect detection results, especially concentrated on square pads and round pads. In view of this, a method for detecting circuit board defects is urgently needed. Summary of the invention

[0004] In view of this, the present application provides a circuit board defect detection method, device and storage medium to solve the technical problem of high leakage rate of PCB defect detection results in the prior art.

[0005] In a first aspect, the present application provides a method for detecting defects in a circuit board, the method comprising:

[0006] Acquire an image of a pad area of ​​a circuit board to be inspected, identify an actual contour of the pad in the image, and generate a virtual contour corresponding to the actual contour;

[0007] Generate a visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper;

[0008] The visual inspection results and AI inspection results of the pad are integrated to obtain the complete inspection results of the pad.

[0009] Preferably, generating a virtual contour corresponding to the actual contour comprises:

[0010] Determine the type and specification of the pad according to the actual contour;

[0011] Based on the type of the pad and the specification of the pad, a virtual outline of a pad standard is fitted on the actual outline.

[0012] Preferably, fitting a virtual contour of a pad standard onto the actual contour comprises:

[0013] If the actual contour is partially missing or blurred, the virtual contour is aligned and fitted with the actual contour according to the type and specification of the pad and the features of the identified clear part.

[0014] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0015] Based on the actual contour and the virtual contour, the deformation width of the pad is calculated, and based on a preset first threshold, a visual inspection result of the pad is generated.

[0016] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0017] The deformation length of the pad is calculated based on the actual contour and the virtual contour, and the visual inspection result of the pad is generated based on a preset second threshold.

[0018] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0019] Based on the actual contour and the virtual contour, the deformation area of ​​the pad is calculated, and based on a preset third threshold, a visual inspection result of the pad is generated.

[0020] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0021] Based on the actual contour and the virtual contour, the total deformation area ratio of the pad is calculated, and based on a preset fourth threshold, a visual inspection result of the pad is generated.

[0022] In a second aspect, the present application provides a circuit board defect detection device, the device comprising:

[0023] An acquisition module, used to acquire an image of a pad area of ​​a circuit board to be inspected, identify an actual contour of the pad in the image, and generate a virtual contour corresponding to the actual contour;

[0024] A detection module, used for generating a visual detection result of the pad based on the actual contour, the virtual contour and a preset visual detection caliper;

[0025] The fusion module is used to fuse the visual inspection results and AI inspection results of the pad to obtain the complete inspection results of the pad.

[0026] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor implementing the circuit board defect detection method described in the first aspect by executing the computer instructions.

[0027] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the circuit board defect detection method described in the first aspect is implemented.

[0028] The circuit board defect detection method, device and storage medium provided by the present application have at least the following beneficial effects:

[0029] The technical solution provided by this application can achieve accurate analysis of multiple dimensions from size, shape to contour deviation by acquiring the image of the pad area and identifying the actual contour, generating a virtual contour, and then combining it with a preset visual inspection gauge for detection. Compared with the detection method that relies solely on AI, this can more comprehensively and meticulously discover defects caused by unclean development of the pad, such as slight dimensional deviations, slight deformations, etc., greatly improving the detection accuracy, effectively avoiding the occurrence of missed detection and false detection, and improving the detection efficiency of circuit board defects. It can promptly discover problems in the pad area, thereby analyzing problems in the development link and ensuring the production quality of the circuit board. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the specific implementation methods or the prior art descriptions. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings without creative work. It should be noted that the drawings described below are schematic and should not be understood as any limitation to the present application. In the drawings:

[0031] Figure 1 A schematic diagram showing a circuit board defect detection method in one embodiment of the present application;

[0032] Figure 2 A schematic diagram of a circular pad on a circuit board is shown;

[0033] Figure 3 A schematic diagram of a qualified pad is shown;

[0034] Figure 4 A schematic diagram of a concave deformation pad is shown;

[0035] Figure 5 A schematic diagram of a deformable pad is shown;

[0036] Figure 6 A schematic diagram showing a circuit board defect detection device in one embodiment of the present application;

[0037] Figure 7 A schematic diagram of a computer device in one embodiment of the present application is shown. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0039] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0040] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, it can also be the internal connection of two components, it can be a wireless connection, or it can be a wired connection. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0041] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0042] Although the processes described below include multiple operations that appear in a specific order, it should be clearly understood that these processes may also include more or fewer operations, and these operations may be performed sequentially or in parallel.

[0043] Example 1

[0044] The circuit board defect detection method provided in this application can be applied to the application scenario of PCB visual defect detection. When using AI models to identify defects on PCB, especially on square pads and round pads, the defect features caused by unclean development are not obvious. The above-mentioned inconspicuousness is relative to obvious defects such as scratches, oil stains, dust, and incompleteness. Please refer to Figure 2 As shown in the figure, it manifests itself as slight convex or concave deformation, which AI cannot effectively detect, resulting in a high miss rate of defect detection results.

[0045] In view of this, please refer to Figure 1 A method for detecting circuit board defects provided in one embodiment of the present application may include the following steps.

[0046] S101, acquiring an image of a pad area of ​​a circuit board to be inspected, identifying an actual contour of the pad in the image, and generating a virtual contour corresponding to the actual contour.

[0047] Specifically, a high-precision industrial camera is used to capture images of the circuit board. This embodiment mainly detects defects caused by unclean development, so the acquired image is an image of the pad area of ​​the circuit board.

[0048] Before the actual contour of the pad is recognized, the acquired original image usually needs to be preprocessed. This includes denoising, which removes the noise generated by camera noise, ambient light interference and other factors in the image. Filtering algorithms such as median filtering and Gaussian filtering can be used; grayscale processing, which converts color images into grayscale images to simplify subsequent image processing operations, because in many contour recognition algorithms, grayscale images are easier to operate; and contrast enhancement processing, which adjusts the contrast of the image to make the edge of the pad and other features more prominent and easier to identify. Use edge detection algorithms to identify the actual contour of the pad. Common edge detection algorithms include Sobel algorithm and Canny algorithm. The Sobel algorithm determines the position of the edge by calculating the gradient of each pixel in the image. It is more sensitive to edge detection in the horizontal and vertical directions; the Canny algorithm is a more advanced multi-stage edge detection algorithm. It first smoothes the image to reduce the impact of noise, then determines possible edge points by calculating the gradient amplitude and direction, and finally accurately extracts the edge through non-maximum suppression and double threshold processing. These algorithms can identify the position of the pad edge in the image, thereby outlining the actual contour of the pad.

[0049] After identifying the actual contour of the pad, it is necessary to generate a corresponding virtual contour. This is usually achieved using a contour fitting algorithm. For example, for some pads with relatively regular shapes (such as circles, squares, etc.), algorithms such as the least squares fitting method can be used. The actual contour of the pad is approximately circular, and an ideal circle closest to the actual contour can be fitted as a virtual contour through the least squares method; for rectangular pads, a standard rectangular virtual contour can be fitted. These fitting algorithms will calculate the virtual contour parameters that can best describe the shape of the contour based on the point coordinate data on the actual contour, such as the center coordinates and radius of the circle, the coordinates of the four vertices of the rectangle, etc. Of course, these parameters can also determine the type and specifications of the pad based on the coordinates of the actual contour of the identified pad. In simple terms, the corresponding actual design data of the pad on the PCB is found, including the type and specifications of the pad, based on the coordinates of the actual contour. The center of the generated virtual contour is consistent with the center of the actual contour. See Figure 3 As shown, Figure 3 There is no defect caused by unclean development on the pad in the figure. In this case, the actual contour and the virtual contour coincide.

[0050] In this method, the actual contour of the pad is identified and a virtual contour corresponding to the actual contour is generated, which is a supplement to AI detection. Compared with the method of performing pixel ratio detection between the detection image and the standard image, or the method of aligning and mapping the detection image with the design drawing and then performing detection, this method can quickly and accurately find the target object that needs additional detection, namely the pad, and can generate the actual contour of the target object and the corresponding virtual contour. The virtual contour is also a defect-free contour, so that the two contours can be used to subsequently identify whether the target object has defects caused by unclean development, thereby avoiding the occurrence of AI detection misses.

[0051] S102: Generate a visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper.

[0052] Specifically, the preset visual inspection gauge is formulated based on the design standards and quality requirements of the pad. It contains a series of key parameters, such as the standard size range of the pad (including length, width, area, etc., which can also be determined according to the shape of the pad), the allowable contour deviation range (in pixels or actual physical dimensions, which specifies the acceptable difference between the actual contour and the standard virtual contour), and the shape regularity requirements (for example, for circular pads, the maximum value of the roundness deviation is specified; for rectangular pads, the allowable fluctuation range of the side length ratio is specified, etc.).

[0053] In this embodiment, the preset visual inspection caliper is a parameter set based on the deviation value of the actual contour and the virtual contour. For example, for a circular pad, the actual contour is relative to the virtual contour, and there is a convex deformation on the actual contour, then the width, length, area, etc. of the convex deformation need to be calculated. The width of the convex deformation is the difference between the maximum radius of the actual contour convex deformation from the center of the virtual contour and the radius of the virtual contour. The length of the convex deformation is the length of the virtual contour between the two intersection points of the actual contour and the virtual contour caused by the convex deformation.

[0054] In this embodiment, when determining whether the pad is qualified or not, if all indicators of the contour deviation between the actual contour and the virtual contour are within the range specified by the visual inspection caliper, then the pad is determined to be a qualified product. Otherwise, the pad is determined to have defects caused by unclean development.

[0055] S103, integrating the visual inspection result and the AI ​​inspection result of the pad to obtain a complete inspection result of the pad.

[0056] Specifically, AI detection usually uses deep learning algorithms, such as convolutional neural networks (CNNs). It can extract and analyze features of pad images, not only limited to contour-related features, but also dig out some deep-level features that may be difficult to capture directly, such as the texture features of the pad surface, the subtle differences of the pad under different lighting conditions, etc. AI detection can be trained with a large amount of labeled pad image data, so that it can identify various normal and abnormal states of the pad, and then make judgments on the newly input pad images to be detected, and give detection results on whether the pad has defects and what kind of defects may exist. Through the method of logical judgment and fusion, the specific defect types pointed out in the visual inspection results and AI inspection results are analyzed separately. If the visual inspection results point out that there are defects caused by unclean development, such as protrusions, depressions, dimensional deviations, etc., and the AI ​​inspection results do not mention them, the visual inspection results and AI inspection results are integrated and output to avoid the loss of AI inspection results.

[0057] In this embodiment, by acquiring the pad area image and identifying the actual contour, generating a virtual contour, and then combining the preset visual inspection gauge for detection, accurate analysis of multiple dimensions from size, shape to contour deviation can be achieved. Compared with the detection method that relies solely on AI, this can more comprehensively and meticulously discover the defects caused by unclean development of the pad, such as slight dimensional deviations, slight deformations, etc., greatly improving the detection accuracy, effectively avoiding the occurrence of missed detection and false detection, and improving the detection efficiency of circuit board defects. Problems in the pad area can be discovered in a timely manner, thereby analyzing problems in the development link and ensuring the production quality of the circuit board.

[0058] Preferably, generating a virtual contour corresponding to the actual contour comprises:

[0059] Determine the type and specification of the pad according to the actual contour;

[0060] Based on the type of the pad and the specification of the pad, a virtual outline of a pad standard is fitted on the actual outline.

[0061] In this embodiment, the type and specification of the pad are determined, that is, according to the actual contour, the corresponding actual design data of the pad is found on the PCB, including the type and specification of the pad. For example, a circular pad with a radius of 0.5 mm. A corresponding virtual contour is generated according to the type and specification of the pad, for example, a circular virtual contour with a radius of 0.5 mm is generated. The virtual contour of the pad is fitted on the actual contour, that is, the generated virtual contour is fitted with the actual contour, for example, the center of the virtual contour is overlapped with the center of the actual contour. The effect after fitting is shown in FIG. Figure 3-Figure 5 shown.

[0062] In this embodiment, by accurately determining the type and specification of the pad, a basis can be provided for the subsequent generation of a standard virtual contour that accurately matches it. For example, for the actual pad contours under different circumstances, a standard virtual contour can be accurately generated, avoiding the inaccuracy of the virtual contour generated by using a single parameter, which leads to deviations in the subsequent detection results. For example, when the actual pad has a uniform unclean development defect as a whole, in simple terms, the pad is one circle larger or one circle smaller as a whole. If only the actual contour is used to fit and generate a virtual contour, the radius of the virtual contour is consistent with the actual contour radius, rather than consistent with the standard radius specification. In this case, the result of using the virtual contour to detect the pad is bound to be inaccurate. By accurately determining the standard virtual contour and the actual contour fitted by the type and specification of the pad, in the subsequent detection and analysis, the subtle differences between the actual pad and the standard can be more finely found, and the slight deviations and deformations in size can be more accurately detected. The detection accuracy is improved, the accuracy of the subsequent detection results is guaranteed, the occurrence of missed detection and false detection is effectively avoided, and the detection efficiency of circuit board defects is improved.

[0063] Preferably, fitting a virtual contour of a pad standard onto the actual contour comprises:

[0064] If the actual contour is partially missing or blurred, the virtual contour is aligned and fitted with the actual contour according to the type and specification of the pad and the features of the identified clear part.

[0065] In this embodiment, when the actual contour is partially missing or blurred, such as a circular pad that is smaller as a whole and has deformations in many places, it is necessary to calculate the center of the circle based on the type of pad, i.e., a circle, and based on the features of the clear parts of the actual contour, such as multiple segments of local clear arc contours, generate a virtual contour based on the pad type and pad specifications (here it can be the radius), and then coincide the center of the virtual contour with the calculated center, so as to fit a standard virtual contour on the actual contour.

[0066] In this embodiment, even if the actual contour is partially missing or blurred, enough valuable information can be extracted to perform alignment and fitting operations based on the type and specifications of the pad and the characteristics of the clear part. For example, for a pad of known circular type, if part of the circumference is blurred, the center position can still be determined more accurately based on the clearly identified part of the circumference and the known diameter specifications and other information, and the shape and size of the entire pad can be judged to meet the requirements by aligning and fitting with the standard virtual contour (generated based on type and specifications). This method ensures that the detection accuracy will not drop significantly when faced with partially unclear actual contours. It ensures detection integrity, maintains detection accuracy, and further improves detection efficiency.

[0067] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0068] Based on the actual contour and the virtual contour, the deformation width of the pad is calculated, and based on a preset first threshold, a visual inspection result of the pad is generated.

[0069] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0070] The deformation length of the pad is calculated based on the actual contour and the virtual contour, and the visual inspection result of the pad is generated based on a preset second threshold.

[0071] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0072] Based on the actual contour and the virtual contour, the deformation area of ​​the pad is calculated, and based on a preset third threshold, a visual inspection result of the pad is generated.

[0073] Preferably, the generating of the visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper comprises:

[0074] Based on the actual contour and the virtual contour, the total deformation area ratio of the pad is calculated, and based on a preset fourth threshold, a visual inspection result of the pad is generated.

[0075] In this embodiment, the preset threshold value is determined based on factors such as the design standard of the pad, the production process requirements, and previous detection experience. This threshold value defines the maximum acceptable degree of deformation of the pad. For example, for a circuit board with high precision requirements, the threshold value of the pad deformation width that may be allowed may be set relatively small. The setting of the threshold value usually needs to be continuously optimized through multiple tests and analysis of actual detection data to ensure that it can accurately reflect the quality status of the pad, so as to ensure that the pad with problems is detected, and not to misjudge too many qualified pads due to the excessively strict threshold value.

[0076] See also Figure 4 As shown, Figure 4 The convex part of the pad, the deformation width of the pad is the difference between the maximum distance from the actual contour to the center of the circle and the radius of the virtual contour, the deformation length of the pad is the length of the virtual contour at the convex part, and the deformation area of ​​the pad is the area where the actual contour protrudes from the virtual contour. In this embodiment, the width, length and area of ​​the deformation must be within the corresponding threshold range to be confirmed as qualified, otherwise they are all unqualified. It should be noted that the concave deformation of the pad should be understood by analogy, and will not be repeated here.

[0077] See also Figure 5 As shown, Figure 5 The deformation includes both convexities and concavities. In this case, it is necessary to judge the convex and concave parts separately, and to judge them as a whole. The overall judgment is that the length of the deformation is the sum of the length of the virtual contour of the convexity and the length of the virtual contour of the concaveity, and the area of ​​the deformation is the sum of the area of ​​the convex virtual contour of the convexity and the area of ​​the concaveity that is missing from the virtual contour.

[0078] In this embodiment, when there are multiple deformations on the pad, it is necessary to calculate the total area of ​​all deformations, and based on the ratio of the total area of ​​the deformation to the virtual contour area, combined with the preset fourth threshold, determine whether the pad is qualified and whether there are defects caused by unclean development.

[0079] In this embodiment, by calculating the deformation parameters of the pad, the deformation of the actual pad contour relative to the standard virtual contour can be quantified. This makes it more accurate to judge whether there are defects caused by unclean development, effectively avoids misjudgment or missed judgment, and improves the detection efficiency. By presenting the deformation degree of the pad with specific parameters, it is possible to clearly and accurately understand the changes in the shape of the pad, so as to more accurately understand the problems existing in the development station, carry out targeted optimization, improve the stability and reliability of the process, and improve the overall production quality of the circuit board.

[0080] Example 2

[0081] This embodiment provides a circuit board defect detection device, which is described by applying it to the circuit board defect detection method provided in the above embodiment 1. Figure 6 As shown, a circuit board defect detection device provided in one embodiment of the present application may include the following multiple modules.

[0082] An acquisition module, used to acquire an image of a pad area of ​​a circuit board to be inspected, identify an actual contour of the pad in the image, and generate a virtual contour corresponding to the actual contour;

[0083] A detection module, used for generating a visual detection result of the pad based on the actual contour, the virtual contour and a preset visual detection caliper;

[0084] The fusion module is used to fuse the visual inspection results and AI inspection results of the pad to obtain the complete inspection results of the pad.

[0085] The circuit board defect detection device provided in the embodiment of the present application can be applied to the circuit board defect detection method provided in the above-mentioned embodiment 1. For relevant details, refer to the above-mentioned method embodiment. The implementation principle and technical effect are similar and will not be repeated here.

[0086] It should be noted that: the circuit board defect detection device provided in the embodiment of the present application only uses the division of the above-mentioned functional modules / functional units as an example when performing circuit board defect detection. In actual applications, the above-mentioned functions can be assigned to different functional modules / functional units as needed, that is, the internal structure of the circuit board defect detection device is divided into different functional modules / functional units to complete all or part of the functions described above. In addition, the implementation method of the circuit board defect detection method provided in the above method embodiment 1 and the implementation method of the circuit board defect detection device provided in this embodiment 2 belong to the same concept. The specific implementation process of the circuit board defect detection device provided in this embodiment 2 is detailed in the above method embodiment 1, and will not be repeated here.

[0087] Example 3

[0088] See also Figure 7 As shown, one embodiment of the present application also provides a computer device, which can be a computer device such as a desktop computer, a laptop computer, a handheld computer, and a cloud server. The computer device can include, but is not limited to, a processor and a memory. The processor and the memory are connected via a bus or other means.

[0089] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, graphics processing units (GPU), embedded neural network processors (NPU) or other dedicated deep learning coprocessors, discrete gate or transistor logic devices, discrete hardware components and other chips, or a combination of the above chips.

[0090] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as program instructions / modules corresponding to the methods in the above-mentioned embodiments of the present application. The processor executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions and modules stored in the memory, that is, implementing the methods in the above-mentioned method embodiments.

[0091] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0092] One embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program. When the computer program is executed by a processor, the method in the above method embodiment is implemented.

[0093] Those skilled in the art can understand that the implementation of all or part of the processes in the above-mentioned implementation methods of the present application can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the implementation methods of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0094] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, 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.

[0095] Although the embodiments of the present application are described in conjunction with the accompanying drawings, they cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, other different forms of changes or modifications can be made on the basis of the above description without departing from the concept of the present application. It is not necessary and impossible to list all the embodiments here. The obvious changes or modifications derived from this are still within the scope of protection created by the present application. Therefore, the scope of protection of the patent application shall be based on the attached claims.

Claims

1. A method for detecting circuit board defects, characterized in that: The method comprises: Acquire an image of a pad area of ​​a circuit board to be inspected, identify an actual contour of the pad in the image, and generate a virtual contour corresponding to the actual contour; Generate a visual inspection result of the pad based on the actual contour, the virtual contour and a preset visual inspection caliper; The visual inspection results and AI inspection results of the pad are integrated to obtain the complete inspection results of the pad.

2. The circuit board defect detection method according to claim 1, characterized in that: The generating of a virtual contour corresponding to the actual contour comprises: Determine the type and specification of the pad according to the actual contour; Based on the type of the pad and the specification of the pad, a virtual outline of a pad standard is fitted on the actual outline.

3. The circuit board defect detection method according to claim 1, characterized in that: Fitting a virtual outline of a pad standard on the actual outline includes: If the actual contour is partially missing or blurred, the virtual contour is aligned and fitted with the actual contour according to the type and specification of the pad and the features of the identified clear part.

4. The circuit board defect detection method according to claim 1, characterized in that: The generating of the visual inspection result of the pad based on the actual contour, the virtual contour and the preset visual inspection caliper comprises: Based on the actual contour and the virtual contour, the deformation width of the pad is calculated, and based on a preset first threshold, a visual inspection result of the pad is generated.

5. The circuit board defect detection method according to claim 4, characterized in that: The generating of the visual inspection result of the pad based on the actual contour, the virtual contour and the preset visual inspection caliper comprises: The deformation length of the pad is calculated based on the actual contour and the virtual contour, and the visual inspection result of the pad is generated based on a preset second threshold.

6. The circuit board defect detection method according to claim 5, characterized in that: The generating of the visual inspection result of the pad based on the actual contour, the virtual contour and the preset visual inspection caliper comprises: Based on the actual contour and the virtual contour, the deformation area of ​​the pad is calculated, and based on a preset third threshold, a visual inspection result of the pad is generated.

7. The circuit board defect detection method according to claim 6, characterized in that: The generating of the visual inspection result of the pad based on the actual contour, the virtual contour and the preset visual inspection caliper comprises: Based on the actual contour and the virtual contour, the total deformation area ratio of the pad is calculated, and based on a preset fourth threshold, a visual inspection result of the pad is generated.

8. A circuit board defect detection device, characterized in that: include: An acquisition module, used to acquire an image of a pad area of ​​a circuit board to be inspected, identify an actual contour of the pad in the image, and generate a virtual contour corresponding to the actual contour; A detection module, used for generating a visual detection result of the pad based on the actual contour, the virtual contour and a preset visual detection caliper; The fusion module is used to fuse the visual inspection results and AI inspection results of the pad to obtain the complete inspection results of the pad.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor implements the circuit board defect detection method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the circuit board defect detection method according to any one of claims 1 to 7 is implemented.

Images