Solder joint inspection method, apparatus, and storage medium

Abstract

The present disclosure provides a solder joint detection method, device and storage medium, the solder joint detection method comprising: obtaining a target image of a circuit board to be detected, the solder joint and the background in the target image showing different effects, the different effects being different colors and / or different shadows; inputting the target image into a solder joint detection model for solder joint detection to obtain a first detection result output by the solder joint detection model, the first detection result including whether the solder joint contained in the circuit board to be detected is missing. The present disclosure can accurately detect whether the circuit board to be detected is missing.

Description

Technical Field

[0001] This disclosure relates to the field of solder joint detection technology, and in particular to a solder joint detection method, equipment and storage medium. Background Technology

[0002] To ensure the secure mounting of electronic components on the circuit board, each pin of the component is soldered to form a solder joint. A good solder joint should completely cover the hole and there should be no adhesion between the solder joints.

[0003] In related technologies, to ensure a high yield rate, Automated Optical Inspection (AOI) technology is used to detect whether there are any missing solder joints. Specifically, a camera is controlled to automatically scan the circuit board to acquire an image of the board; then, the acquired image is transmitted to a processor, which compares the solder joints in the image with the corresponding qualified solder joints on the circuit board, thereby detecting missing solder joint defects on the circuit board.

[0004] However, the inventors discovered that when using AOI technology to inspect solder joints on circuit boards, there is a problem of misjudging missing solder joints. Summary of the Invention

[0005] This disclosure provides a solder joint inspection method, apparatus, and storage medium to address the problem of misjudgment of missing solder joints during current solder joint inspection of circuit boards.

[0006] The first aspect of this disclosure provides a solder joint detection method, comprising: acquiring a target image of a circuit board to be detected, wherein the solder joints and the background in the target image are presented with different effects, the different effects being different colors and / or different shadows; inputting the target image into a solder joint detection model for solder joint detection, and obtaining a first detection result output by the solder joint detection model, the first detection result including whether there are any missing solder joints in the solder joints contained in the circuit board to be detected.

[0007] A second aspect of this disclosure provides a solder joint detection device, comprising:

[0008] The acquisition module is used to acquire the target image of the circuit board to be inspected. The solder joints and background in the target image are presented with different effects, which are different in color and / or shadow.

[0009] The detection module is used to input the target image into the solder joint detection model to perform solder joint detection and obtain the first detection result output by the solder joint detection model. The first detection result includes whether there are any missing solder joints in the circuit board to be detected.

[0010] A third aspect of this disclosure provides an electronic device, including: a memory and a processor;

[0011] Memory, used to store executable instructions;

[0012] The processor is configured to execute executable instructions to implement the solder joint detection method as described in the first aspect.

[0013] A fourth aspect of this disclosure provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the solder joint detection method of the first aspect.

[0014] The fifth aspect of this disclosure provides a computer program product, the program product comprising: a computer program stored in a readable storage medium, at least one processor of an electronic device being able to read the computer program from the readable storage medium, and the at least one processor executing the computer program causing the electronic device to perform the solder joint detection method of the first aspect.

[0015] The solder joint detection method, device, and storage medium provided in this disclosure address the issue that solder joints and the background appear differently in the target image of the circuit board to be inspected. Therefore, when the target image enters the solder joint detection model for solder joint detection, the model can more accurately identify the position and size of the solder joints, thereby more accurately determining whether there are any missing solder joints on the circuit board to be inspected. This solves the problem of misjudging missing solder joints during circuit board solder joint detection and improves the detection accuracy of the circuit board to be inspected. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this disclosure and form part of this disclosure, illustrate exemplary embodiments of the present disclosure and are used to explain the disclosure, but do not constitute an undue limitation of the disclosure. In the drawings:

[0017] Figure 1 An application scenario diagram of a solder joint detection method provided for an exemplary embodiment of this disclosure;

[0018] Figure 2 A flowchart illustrating the steps of a solder joint detection method provided as an exemplary embodiment of this disclosure;

[0019] Figure 3 Example diagram of illumination light reflection provided for exemplary embodiments of this disclosure;

[0020] Figure 4a Example diagram of a good solder joint in an acquired image provided for an exemplary embodiment of this disclosure;

[0021] Figure 4b Example diagram of missing solder joints in the acquired images provided for exemplary embodiments of this disclosure;

[0022] Figure 5Registration correction example diagrams provided for exemplary embodiments of this disclosure;

[0023] Figure 6 A flowchart illustrating the steps of a weld joint bridging detection method provided as an exemplary embodiment of this disclosure;

[0024] Figure 7 Example diagram of solder joint location provided for exemplary embodiments of this disclosure;

[0025] Figure 8 A flowchart illustrating the steps of another solder joint detection method provided as an exemplary embodiment of this disclosure;

[0026] Figure 9 Structural block diagram of a solder joint detection device provided for an exemplary embodiment of this disclosure;

[0027] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an example embodiment of this disclosure. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this disclosure clearer, the technical solutions of this disclosure will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0029] A PCB (Printed Circuit Board) is a support structure for electronic components, enabling electrical connections between them in a circuit. Electronic components mounted on a PCB have at least one pin. To secure the component to the PCB, the pin passes through holes in the PCB and is soldered on the other side to form a solder joint. A good solder joint should completely cover the hole through which the corresponding pin passes and should not adhere to other solder joints.

[0030] To ensure a high yield rate, the solder joints on the PCB undergo quality inspection. Common inspection methods include: a) On-site operators inspect the solder joints using a magnifying glass or by visual inspection, relying on experience to determine the PCB's quality. However, this method is purely experience-based, inefficient, and prone to missed defects. b) Fully automated circuit board testing lines integrate ICT (In-Circuit Test) and FCT (Functional Circuit Test) into an automated production line. ICT testing involves testing the resistance, capacitance, inductance, and continuity of each component on the product when it is not powered on, checking for normal component status. FCT testing involves powering on the product and testing its normal operating parameters, primarily used to check the circuit board's functionality. This method requires customized probes and probe beds for each board type, which must contact the soldered components to function, and also necessitates the design of specific test points on the circuit board. Furthermore, after detecting a problem, it's difficult to quickly locate the problematic solder joint, and there's a possibility that cases where component leads are in contact with the PCB but are missing solder joints may go undetected. c. Using Automated Optical Inspection (AOI) technology to detect missing solder joints on the PCB. AOI technology works by using a camera to output quantitative grayscale values ​​of the reflected light from the inspected object, then comparing these values ​​with the grayscale values ​​of a standard image to analyze and classify defects. However, when acquiring PCB images using AOI technology, there may be areas where grayscale or shadows are not clearly defined. These areas are difficult to identify, leading to false positives for missing solder joints and affecting the accuracy of the inspection.

[0031] Based on the above problems, the solder joint detection method, device and storage medium provided in this disclosure make different components in the acquired image present different effects, so that the acquired image can be better distinguished and identified by the system, thereby obtaining more accurate solder joint detection results.

[0032] In this embodiment, the solder joint detection method can be implemented using a cloud computing system. Furthermore, the server executing the solder joint detection method can be a cloud server, leveraging the advantages of cloud resources to run various algorithms; however, the solder joint detection method can also be applied to conventional servers or server arrays, and is not limited thereto.

[0033] Furthermore, one application scenario of this disclosure embodiment is as follows: Figure 1 As shown, Figure 1The system includes a camera 11, a processing center 12, a light source 13, and a circuit board 14. Electronic components (not shown in the figure) are mounted on the circuit board 14, including but not limited to resistors, capacitors, inductors, and transistors. Illumination light generated by the light source 13 shines on the circuit board 14. The camera 11 is positioned above the circuit board 14, and the reflected light from the circuit board 14 is captured by the camera 11 to form an image. The image is transmitted to the processing center 12 for processing to obtain the solder joint detection results.

[0034] also, Figure 1 This is merely one exemplary application scenario, and the embodiments disclosed herein can be applied to any welding joint inspection scenario. The embodiments disclosed herein do not limit the specific application scenario.

[0035] Figure 2 A flowchart illustrating the steps of a solder joint inspection method provided for an exemplary embodiment of this disclosure. (Refer to...) Figure 2 The specific steps of the solder joint inspection method include:

[0036] S201, acquire the target image of the circuit board to be inspected, in which the solder joints and background are presented with different effects.

[0037] The different effects mentioned above can be presented in various ways, such as using different colors or different shadows.

[0038] The target image can be an image captured by a camera and uploaded to the detection device. The camera can be a separate device or included in the detection device.

[0039] In some embodiments, the detection device is also equipped with an infrared sensor, which triggers a camera to take a picture when the circuit board to be detected passes by the infrared sensor.

[0040] S202, input the target image into the solder joint detection model to perform solder joint detection, and obtain the first detection result output by the solder joint detection model.

[0041] The first detection result includes whether there are any missing solder joints on the circuit board under test. Optionally, if there are missing solder joints on the circuit board under test, "NG" can be output, and the location of the missing solder joint can be marked on the image for later resoldering; if there are no missing solder joints on the circuit board under test, "OK" can be output, and the circuit board under test can proceed to the next process. The solder joint detection model can be a model obtained through deep learning.

[0042] In some embodiments, the solder joint detection model can be a model obtained by training a ResNet18 network using sample images. A ResNet18 (Deep Residual Network) refers to an 18-layer deep residual network. The depth of the network is crucial to the model's performance, and because the accuracy of the ResNet network is not affected by increasing its depth, it can represent more layers of the model.

[0043] In this embodiment, since the solder joints and background appear differently in the target image of the circuit board to be inspected, the solder joint detection model can more accurately identify the position and size of the solder joints when the target image enters the solder joint detection model for solder joint detection. This allows for a more accurate determination of whether there are any missing solder joints on the circuit board to be inspected. This solves the problem of misjudgment of missing solder joints during circuit board solder joint detection and improves the detection accuracy of the circuit board to be inspected.

[0044] Optionally, the training method for the aforementioned solder joint detection model may include: inputting sample images into a ResNet18 network to obtain output results, comparing the output results with the correct results, calculating the error, and then performing further learning based on the error. The loss function can express the difference between the output results obtained through the ResNet18 network and the correct results.

[0045] The loss function can be expressed as follows:

[0046]

[0047] In the above formula, z i Let z be the output value of the i-th node, C be the number of output nodes (i.e., the number of categories), and z be c×1. The Softmax function can be used to convert the output values ​​of a multi-class network into a probability distribution in the range [0, 1] with a sum of 1. The dimension of the probability distribution is the same as z.

[0048] Optionally, to make the solder joints and background appear differently in the target image, the image acquisition device can be controlled to acquire an image of the circuit board under illumination at an angle of 20° to 40° with respect to the circuit board under inspection, thus obtaining the target image. For example, the illumination light can form an angle of 20° with the circuit board under inspection, or an angle of 30° with the circuit board under inspection, or an angle of 40° with the circuit board under inspection.

[0049] Because the solder joints and the background are not on the same plane, the reflected light is also different. The solder joints will be brighter.

[0050] In some embodiments, to make the different effects of solder joints and background in the target image more prominent, the image acquisition device is controlled to acquire the image of the circuit board to be inspected when different colored illumination lights form an angle with the circuit board to be inspected, and the size of the angle is different, so as to obtain the target image.

[0051] Because the solder joints are prominent relative to the background, when illuminated by light of different angles and colors, the principle of light reflection will cause different color effects to appear on the solder joints and the background. (Refer to...) Figure 3 There are two light sources 32 emitting different illumination lights. Taking the plane of the circuit board under test as a reference, the illumination light with a smaller tilt angle (refer to...) Figure 3 The light rays (shown by the dotted lines in the image) will shine on the solder joint 33 and then be reflected by the lens and captured by the camera 31. Light rays with a large tilt angle (see reference) Figure 3 The light rays (shown by the solid lines in the image) and those with a small angle of inclination can both illuminate the background 34, resulting in different colors appearing in the camera's captured image for different feature areas. In the captured image, the solder joint's color is the color of the light with a small angle of inclination, while the background color is either the color of the light with a small angle of inclination or a mixture of the light with a large angle of inclination and the light with a small angle of inclination. If a pin is missing a solder joint—that is, if it should have been soldered but wasn't—the missing area will appear black because of the hole through which the pin passed. Therefore, color can be used to distinguish the solder joint, the background, and the location of the missing solder joint. Furthermore, some pins may protrude beyond the solder joint; under the aforementioned illumination, the pin's color will also be the same as the background color.

[0052] It is important to note that, in order to ensure accurate differentiation, the colors of the two types of illumination light should be sufficiently distinct. In addition, the angle of the illumination light should be adjusted based on the actual situation to ensure that the reflected light can be received by the camera.

[0053] Furthermore, optionally, the image acquisition device can be controlled to acquire an image of the circuit board under the illumination of red light forming a first angle with the circuit board under test and blue light forming a second angle with the circuit board under test, thereby obtaining a target image.

[0054] The first included angle is between 50° and 70°, and the second included angle is between 20° and 40°. For example, the first included angle can be 50°, 60°, or 70°, and the second included angle can be 20°, 30°, or 40°.

[0055] For example, a blue light at a 60° angle and a red light at a 30° angle to the circuit board under test can be used to illuminate the board. Under these lighting conditions, the blue light shining on the solder joints reflects the light to the lens of the color camera. The blue and red light simultaneously illuminate the background and pins. Because the reflection angles of the background and pins are different from those of the solder joints, different feature areas captured by the color camera will have different colors of light. The solder joints will appear blue, and the background and pins will appear red or a purplish-red mixture of blue and red. Figure 4a and Figure 4b In the diagram, the hexagonal shading represents blue, the square shading represents red, the diagonal shading represents magenta, and the blank areas are black. If the solder joints are good, refer to... Figure 4a The image shows that the area with more blue features indicates a solder joint; if there is a missed solder joint, refer to... Figure 4b The blue area at this location lacks the black features, while the pin and background features remain red or purplish-red.

[0056] In some embodiments, the target image is first registered and corrected to obtain a corrected image, and then the corrected image is input into the solder joint detection model to detect solder joints, thereby obtaining the first detection result output by the solder joint detection model.

[0057] Registration in registration correction refers to finding more than four similar objects on the target image that match the standard image. Figure 1 Similar marker points. Correction refers to a series of changes to an image, such as rotation, translation, and scaling, to transform it into a standard image. This process requires taking an image as the standard image. The target image can be considered as the image generated by scanning the printed standard image. Therefore, the target image can undergo non-rigid changes such as rotation, scaling, and translation to become an image with the same height, angle, and other parameters as the standard image. The registered and corrected target image is easier for the solder joint detection model to detect.

[0058] For example, refer to Figure 5 If the target image captured by the camera is the original image, it needs to be rotated by a certain angle compared to the standard image. After the registration and correction algorithm, the target image will be rotated until it is at the same angle as the standard image.

[0059] Optionally, since there are many solder joints in the circuit board to be inspected, the target image is first sliced ​​to obtain multiple sub-images, each containing at least one solder joint. For each sub-image, the sub-image is input into the solder joint detection model for solder joint detection, and a second detection result corresponding to the sub-image output by the solder joint detection model is obtained. Based on the second detection results corresponding to the multiple sub-images, a first detection result output by the solder joint detection model is determined.

[0060] Accordingly, if any of the second test results shows a missing solder joint, it indicates that the circuit board under test has a missing solder joint and needs to be repaired.

[0061] Generally, when performing solder joint defect detection, the sub-image is cut into smaller segments, down to each individual solder joint.

[0062] It is understood that the registration and image slicing processing of the target image in the above embodiments can be used in combination or separately, and there is no limitation here.

[0063] The above embodiments specifically describe the detection method for missing solder joints on the circuit board to be tested. In addition to missing solder joints, solder joint bridging can also affect the quality of the circuit board. The following embodiments describe the detection method for solder joint bridging on the circuit board to be tested.

[0064] Figure 6 A flowchart illustrating the steps of a solder joint bridging detection method provided for an exemplary embodiment of this disclosure. (Refer to...) Figure 6 The specific steps of the weld joint bridging inspection method include:

[0065] S601, perform blob analysis on the target image to obtain the solder joint location map.

[0066] This step is performed only if the first test result indicates that there are no missing solder joints on the circuit board under test. This is because if solder joint bridging is checked first, and there are missing solder joints later, the resoldered solder joints may also bridging with surrounding solder joints, thus requiring two solder joint bridging checks.

[0067] Blob (binary large object) analysis essentially involves binarizing an image to segment it into foreground and background, then performing connected component detection to obtain blob blocks. It can also be viewed as finding small regions with abrupt changes in grayscale within a smooth area. Solder joints are protruding regions relative to the background, therefore, blob analysis will locate and mark them. For example, a solder joint location map after blob analysis could be shown as follows... Figure 7 As shown.

[0068] S602, based on the solder joint location diagram, determine whether there is solder joint bridging on the circuit board to be tested.

[0069] Optionally, there are several ways to determine whether there is solder joint bridging on the circuit board under test. In the first implementation, the solder joint location map can be manually checked to determine whether solder joint bridging occurs. In another implementation, the solder joint location map can be input into a bridging detection model for bridging detection, and a third detection result output by the bridging detection model can be obtained. The third detection result includes whether there is solder joint bridging on the circuit board under test.

[0070] The target image can also be registered, corrected, and sliced ​​before blob analysis. However, for sliced ​​image processing, since solder joint detection is required, the target image cannot be sliced ​​too small. Generally, it is sliced ​​down to each individual device, and the resulting small image needs to contain multiple solder joints.

[0071] Figure 8 A flowchart illustrating the steps of another solder joint detection method provided as an exemplary embodiment of this disclosure. (Refer to...) Figure 8 The specific steps of the solder joint inspection method include:

[0072] S801, acquire the target image of the circuit board to be inspected.

[0073] The solder joints and background in the target image are presented with different effects, namely different colors and / or different shadows.

[0074] S802, register and correct the target image to obtain the corrected image.

[0075] S803 performs image slicing on the corrected image to obtain multiple first sub-images.

[0076] S804: For each of the multiple first sub-images, input the first sub-image into the solder joint detection model to perform solder joint detection, and obtain the second detection result corresponding to the first sub-image output by the solder joint detection model.

[0077] S805, based on the second detection results corresponding to multiple first sub-images, determine the first detection result output by the solder joint detection model.

[0078] S806, determine whether there is any missing solder joints in the circuit board to be tested, as indicated by the first detection result.

[0079] If yes, proceed to step S807; otherwise, proceed to step S808.

[0080] S807, returned for repair.

[0081] S808 performs image slicing on the corrected image to obtain multiple second sub-images.

[0082] Because the second sub-image needs to be analyzed by blob and further detected for solder joints, the second sub-image is usually cut according to the electronic components, and each second sub-image contains multiple solder joints.

[0083] S809 performs blob analysis on multiple second sub-images to obtain multiple solder joint location maps.

[0084] S810 determines whether there is solder joint bridging on the circuit board under test based on multiple solder joint location diagrams.

[0085] If yes, proceed to step S807; otherwise, it means that the circuit board to be inspected has neither missing solder nor bridging solder, and the solder joint inspection can be terminated to proceed with the next steps.

[0086] In this embodiment of the disclosure, reference is made to Figure 9 In addition to providing a solder joint inspection method, a solder joint inspection device 90 is also provided, including:

[0087] The acquisition module 901 is used to acquire a target image of the circuit board to be inspected. The solder joints and the background in the target image are presented with different effects, which are different colors and / or different shadows.

[0088] The detection module 902 is used to input the target image into the solder joint detection model to perform solder joint detection and obtain the first detection result output by the solder joint detection model. The first detection result includes whether there is any missing solder joint in the solder joints contained in the circuit board to be detected.

[0089] In one optional embodiment, the acquisition module 901 is specifically used to: control the image acquisition device to acquire images of the circuit board to be inspected when different colors of illumination light form an angle with the circuit board to be inspected, and the size of the angle is different, so as to obtain a target image.

[0090] In one optional embodiment, the acquisition module 901 is specifically used to: control the image acquisition device to acquire an image of the circuit board to be tested under the illumination of red light forming a first angle with the circuit board to be tested and blue light forming a second angle with the circuit board to be tested, thereby obtaining a target image; the size of the first angle is 50° to 70°, and the size of the second angle is 20° to 40°.

[0091] In one optional embodiment, the detection module 902 inputs the target image into the solder joint detection model to perform solder joint detection and obtains the first detection result output by the solder joint detection model. Specifically, it is used to: register and correct the target image to obtain a corrected image; input the corrected image into the solder joint detection model to perform solder joint detection and obtain the first detection result output by the solder joint detection model.

[0092] In one optional embodiment, the detection module 902 inputs the target image into the solder joint detection model for solder joint detection, and obtains a first detection result output by the solder joint detection model. Specifically, it is used to: perform image slicing on the target image to obtain multiple sub-images, each sub-image containing at least one solder joint; for each sub-image in the multiple sub-images, input the sub-image into the solder joint detection model for solder joint detection, and obtain a second detection result corresponding to the sub-image output by the solder joint detection model; and determine the first detection result output by the solder joint detection model based on the second detection results corresponding to the multiple sub-images.

[0093] In one optional embodiment, the detection module 902 is further configured to: if the first detection result indicates that there are no missing solder joints in the solder joints contained in the circuit board to be detected, perform blob analysis on the target image to obtain a solder joint location map; and determine whether there are solder joint bridging on the circuit board to be detected based on the solder joint location map.

[0094] In one optional embodiment, the detection module 902 determines whether there is solder joint bridging on the circuit board to be tested based on the solder joint location diagram. Specifically, it is used to: input the solder joint location diagram into the bridging detection model for bridging detection, and obtain a third detection result output by the bridging detection model. The third detection result includes whether there is solder joint bridging on the circuit board to be tested.

[0095] The solder joint detection apparatus provided in this embodiment addresses the issue that, during solder joint detection, the solder joints and background in the target image of the circuit board to be inspected present different effects. Therefore, the solder joint detection model can more accurately identify the position and size of the solder joints, thereby more accurately determining whether there are any missing solder joints on the circuit board to be inspected. This solves the problem of misjudgment of missing solder joints during circuit board solder joint detection and improves the detection accuracy of the circuit board to be inspected.

[0096] Furthermore, in some of the processes described in the above embodiments and accompanying drawings, multiple operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The sequence numbers are merely used to distinguish different operations, and the sequence numbers themselves do not represent any execution order. Additionally, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0097] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an example embodiment of this disclosure. For example... Figure 10 As shown, the electronic device 100 includes a processor 1001 and a memory 1002 communicatively connected to the processor 1001, the memory 1002 storing computer-executable instructions.

[0098] The processor executes computer execution instructions stored in the memory to implement the solder joint detection method provided in any of the above method embodiments. The specific functions and technical effects to be achieved will not be elaborated here.

[0099] This disclosure also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the solder joint detection method provided in any of the above method embodiments.

[0100] This disclosure also provides a computer program product, which includes a computer program stored in a readable storage medium. At least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the solder joint detection method provided in any of the above method embodiments.

[0101] In the embodiments provided in this disclosure, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0102] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0103] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0104] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods of the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is merely an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0106] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0107] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for detecting solder joints, characterized in that, include: Acquire a target image of the circuit board to be inspected, wherein the solder joints and the background in the target image are presented with different effects, the different effects being different colors and / or different shadows; The target image is registered and corrected to obtain a corrected image; The corrected image is subjected to a first image slicing process to obtain a plurality of first sub-images, each of which includes at least one solder joint. For each of the multiple first sub-images, the first sub-image is input into the solder joint detection model for solder joint detection, and a second detection result corresponding to the first sub-image output by the solder joint detection model is obtained. Based on the second detection results corresponding to the multiple first sub-images, a first detection result output by the solder joint detection model is determined. The first detection result includes whether there is any missing solder joint in the solder joints contained in the circuit board to be detected. If the first detection result indicates that there are no missing solder joints in the circuit board to be tested, then the corrected image is subjected to a second image cropping process to obtain multiple second sub-images; each second sub-image contains multiple solder joints; wherein, the second sub-image is larger than the first sub-image; Blob analysis was performed on the multiple second sub-images to obtain multiple solder joint location maps; The multiple solder joint location maps are input into the bridging detection model for bridging detection, and a third detection result is obtained from the bridging detection model. The third detection result includes whether there is bridging solder joint on the circuit board to be tested.

2. The solder joint detection method according to claim 1, characterized in that, The process of acquiring the target image of the circuit board to be inspected includes: Under illumination light forming an angle of 20° to 40° with the circuit board to be inspected, the image acquisition device is controlled to acquire an image of the circuit board to be inspected, thereby obtaining the target image.

3. The solder joint detection method according to claim 1, characterized in that, The process of acquiring the target image of the circuit board to be inspected includes: When different colors of illumination light form angles with the circuit board to be inspected, and the size of the angles is different, the image acquisition device is controlled to acquire images of the circuit board to be inspected, thereby obtaining the target image.

4. The solder joint detection method according to claim 3, characterized in that, When different colors of illumination light form angles with the circuit board to be inspected, and the angles are of different sizes, the image acquisition device is controlled to acquire images of the circuit board to be inspected to obtain the target image, including: Under the illumination of red light forming a first angle with the circuit board to be inspected and blue light forming a second angle with the circuit board to be inspected, the image acquisition device is controlled to acquire an image of the circuit board to be inspected, thereby obtaining the target image; The first included angle is 50° to 70°, and the second included angle is 20° to 40°.

5. A solder joint detection device, characterized in that, include: The acquisition module is used to acquire a target image of the circuit board to be inspected, wherein the solder joints and the background in the target image are presented with different effects, and the different effects are different colors and / or different shadows; The detection module is used to register and correct the target image to obtain a corrected image; The corrected image is subjected to a first image slicing process to obtain a plurality of first sub-images, each of which includes at least one solder joint. For each of the multiple first sub-images, the first sub-image is input into the solder joint detection model for solder joint detection, and the second detection result corresponding to the first sub-image is output by the solder joint detection model. Based on the second detection results corresponding to the multiple first sub-images, the first detection result output by the solder joint detection model is determined. The first detection result includes whether there is any missing solder joint in the solder joints contained in the circuit board to be detected. If the first detection result indicates that there are no missing solder joints in the circuit board to be tested, then the corrected image is subjected to a second image cropping process to obtain multiple second sub-images; each second sub-image contains multiple solder joints; wherein, the second sub-image is larger than the first sub-image; blob analysis is performed on the multiple second sub-images to obtain multiple solder joint location maps; The multiple solder joint location maps are input into the bridging detection model for bridging detection, and a third detection result is obtained from the bridging detection model. The third detection result includes whether there is bridging of solder joints on the circuit board to be tested.

6. An electronic device, characterized in that, include: Memory, processor; The memory is used to store executable instructions; The processor is configured to execute the executable instructions to implement the solder joint detection method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the solder joint detection method as described in any one of claims 1 to 4.

8. A computer program product, characterized in that, It includes a computer program, which, when executed by a processor, is used to implement the solder joint detection method as described in any one of claims 1 to 4.

Images