Welded product quality detection method and device and quality detection equipment

Abstract

The embodiment of the invention provides a welding product quality detection method and device and product inspection equipment. The method is applied to product inspection equipment, and standard images and a welding defect feature description information base are stored in the product inspection equipment. The standard image corresponds to a preset qualified form of the welding product, the standard image comprises a plurality of welding spots, and each welding spot is provided with a defect identifier used for representing the possibility that the welding spot has a specific welding type defect; the method comprises the following steps: acquiring a to-be-processed image; the to-be-processed image is compared with the standard image, and whether all welding features contained in the to-be-processed image are completely consistent with the standard image or not is judged; if not, extracting a feature image; determining the possibility of a specific welding type defect according to a defect identifier corresponding to each welding spot in the feature image; according to the possibility of the specific welding type defect, feature information matched with the possibility is called, the feature image is compared with the feature information, and the actual defect type of the welding product is determined.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of welding technology. More specifically, the embodiments of the present application relate to a welding product quality detection method, a welding product quality detection device and quality inspection equipment. Background Art

[0002] The quality of welded products is directly related to the performance, safety, and reliability of the entire product. Whether welding is done manually or mechanically, there is a certain probability of weld defects such as missing solder, insufficient solder, misaligned soldering, and bridging. These defects can have a serious impact on product reliability, making it particularly important to inspect the weld quality of welded products.

[0003] Existing machine vision-based inspection methods simply compare the image to be inspected with a preset standard image, and determine it as a defect when inconsistency occurs. However, they cannot accurately determine the type of defect and lack in-depth analysis and judgment of the defect.

[0004] In view of this, it is necessary to provide a new technical solution to solve the above technical problems. Summary of the Invention

[0005] The purpose of this application is to provide a new technical solution for welding product quality inspection method, device and quality inspection equipment.

[0006] According to the first aspect of the embodiment of the present application, a method for detecting the quality of welding products is provided. The method is applied to a quality inspection device, wherein the quality inspection device pre-stores a standard image and a welding defect feature description information library; wherein The standard image corresponds to a preset qualified form of the welded product, and the standard image includes a plurality of welds, each weld being provided with a defect identifier, wherein the defect identifier is used to indicate the possibility of a specific welding type defect occurring in the weld; The welding product quality detection method comprises the following steps: Acquire an image to be processed, where the image to be processed is an image of a welding product to be inspected; Comparing the image to be processed with the standard image; Determine whether all welding features contained in the image to be processed are completely consistent with the standard image; if they are completely consistent, determine that the welded product is a qualified product; if they are not completely consistent, extract a feature image based on the image to be processed, wherein the feature image includes images of weld spots that are inconsistent with the welding features in the standard image; Determining the likelihood of a specific welding type defect occurring in each inconsistent weld point based on the defect identifier corresponding to each weld point image in the feature image; Based on the possibility of a specific welding type defect occurring in each inconsistent weld spot, feature information matching the possibility is retrieved from a welding defect feature description information library, and the feature image is compared with the feature information to determine the actual defect type of the welded product.

[0007] Optionally, comparing the image to be processed with the standard image specifically includes: Extract key welding features from the image to be processed and the standard image; Calculate the similarity between the corresponding features in the image to be processed and the standard image.

[0008] Optionally, extracting a feature image based on the image to be processed specifically includes: Calculate the difference area between the image to be processed and the standard image, and use the image in the difference area as the preliminary feature image; The preliminary feature image is processed to obtain the final feature image.

[0009] Optionally, each weld point is provided with a defect mark, specifically including: Collect sample images of welding products with different defect types, annotate and analyze the welds in each sample image, calculate the probability of each weld having various specific welding defects, and set a corresponding defect identifier for each weld based on the statistical results. The defect identifier is expressed as a probability value or level.

[0010] Optionally, obtaining the image to be processed specifically includes: The welded product to be inspected is photographed from multiple preset angles, and the photographed images are preprocessed to obtain images to be processed, wherein the preprocessing includes denoising and contrast enhancement operations.

[0011] Optionally, before obtaining the image to be processed, the following steps are further included: Acquire employee hand motion images, analyze and process the acquired employee hand motion images based on a pre-built motion analysis model, and determine whether the conditions for collecting images of the welding product to be inspected are met based on the analysis results. If so, perform subsequent collection operations. If not, remind the employee to perform hand motion operations.

[0012] Optionally, obtaining the employee's hand motion image specifically includes: Acquire an image of an employee wearing a wristband to prevent electrical breakdown of a welding product, acquire an image of an employee wiping the surface of a welding product, and acquire an image of an employee wiping the surface of a welding product and placing the welding product in the original area.

[0013] Optionally, after determining the actual defect type of the welding product, the method further includes: The inspection results are recorded and stored, including the welding product number, inspection time, defect type and defect location information; and the inspection results are fed back to the production control system.

[0014] According to the second aspect of the embodiment of the present application, a welding product quality detection device is provided. The detection device is applied to a quality inspection device, wherein the quality inspection device pre-stores a standard image and a welding defect feature description information library; wherein The standard image corresponds to a preset qualified form of the welded product, and the standard image includes a plurality of welds, each of which is provided with a defect identifier, wherein the defect identifier is used to indicate the possibility of a specific welding type defect occurring in the weld; The welding product quality detection device comprises: An image acquisition module is used to acquire an image to be processed, wherein the image to be processed is an image of the welded product to be inspected; A quality detection module, configured to compare the image to be processed with the standard image; a judgment module for judging whether all welding features contained in the image to be processed are completely consistent with those in the standard image; if they are completely consistent, the welded product is determined to be a qualified product; if they are not completely consistent, a feature image is extracted from the image to be processed by an extraction module, the feature image including images of weld spots that are inconsistent with the welding features in the standard image; a determination module, determining the possibility of a specific welding type defect occurring in each inconsistent weld point based on the defect identifier corresponding to each weld point image in the feature image; The comparison module retrieves feature information that matches the possibility of a specific welding type defect occurring in each inconsistent weld point from a welding defect feature description information library, compares the feature image with the feature information, and determines the actual defect type of the welded product.

[0015] According to a third aspect of an embodiment of the present application, a quality inspection device is provided, which includes the welding product quality inspection device as described in the second aspect; or includes a processor and a memory, the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute the welding product quality inspection method as described in the first aspect.

[0016] In the technical solution provided in the embodiments of the present application, a welding product quality inspection method is applied to quality inspection equipment, which pre-stores a standard image and a welding defect feature description information library. The standard image corresponds to a preset qualified form of the welding product, and the standard image includes multiple welds, each of which is provided with a defect identifier to indicate the likelihood of a specific welding type defect at that weld. By comparing the collected image of the welding product to be inspected with the standard image, extracting the feature image, and determining the actual defect type of the welding product based on the defect identifier of the weld and the welding defect feature description information library, the quality of the welding product can be more accurately and comprehensively inspected, improving the accuracy and reliability of the inspection, while also increasing the inspection efficiency.

[0017] Other features and advantages of the present specification will become apparent from the following detailed description of exemplary embodiments of the present specification with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the specification and, together with the description, serve to explain the principles of the specification.

[0019] Figure 1 Shown is a flow chart of a welding product quality detection method provided in an embodiment of the present application.

[0020] Figure 2 Shown is a functional module diagram of a welding product quality detection device provided in an embodiment of the present application.

[0021] Figure 3 A block diagram of a quality inspection device provided by an embodiment of the present invention is shown.

[0022] Description of reference numerals: 200, welding product quality detection device; 201, image acquisition module; 202, quality detection module; 203, judgment module; 204, extraction module; 205, determination module; 206, comparison module; 300 , quality inspection equipment; 310 , bus; 320 , processor; 330 , memory; 340 , I / O module; 350 , communication interface. DETAILED DESCRIPTION

[0023] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0024] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0025] Techniques and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the techniques and equipment should be considered part of the specification.

[0026] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0027] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0028] Please refer to Figure 3 , is a block diagram of a quality inspection device 300 provided in an embodiment of the present application. The quality inspection device 300 includes a bus 310, a processor 320, a memory 330, an I / O module 340, and a communication interface 350.

[0029] The bus 310 may be a circuit that interconnects the aforementioned elements and transmits communications (e.g., control messages) between the aforementioned elements. The processor 320 may receive commands from the aforementioned other elements (e.g., the memory 330, the I / O module 340, the communication interface 350, etc.) via the bus 310, interpret the received commands, and perform calculations or data processing based on the interpreted commands.

[0030] The processor 320 can be an integrated circuit chip with signal processing capabilities. The processor 320 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The memory 330 can store commands or data received from the processor 320 or other components (such as the I / O module 340, the communication interface 350, etc.) or commands or data generated by the processor 320 or other components. The memory 330 can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), or an electrically erasable programmable read-only memory (EEPROM). The I / O module 340 can receive commands or data input from the user via input-output means (e.g., sensors, keyboards, touch screens, etc.), and can transmit the received commands or data to the processor 320 or the memory 330 via the bus 310. It is also used to display various information (e.g., multimedia data, text data) received, stored, and processed from the above components, and can display videos, images, data, etc. to the user. The communication interface 350 can be used to communicate signaling or data with other node devices. It is understandable that Figure 3 The structure shown is only a schematic diagram of the structure of the inspection equipment 300. The inspection equipment 300 may also include Figure 3 More or fewer components than shown, or with Figure 3 Different configurations shown. Figure 3 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0031] The above-mentioned quality inspection equipment 300 will be used as the execution body to execute each step of each method provided in the embodiments of this application and achieve corresponding technical effects. <Method Example> An embodiment of the present application provides a method for inspecting the quality of welding products. The method is applied to quality inspection equipment 300. The quality inspection equipment 300 pre-stores a standard image and a library of welding defect feature descriptions. The standard image corresponds to a preset qualified form of the welding product, and the standard image includes multiple welds, each of which is assigned a defect identifier that indicates the likelihood of a specific type of welding defect occurring at that weld. Using this method, the quality inspection equipment can analyze and process the collected images of the welding product to be inspected, accurately determine whether the welded product is qualified, and identify the actual defect type.

[0032] This welding product quality inspection method is implemented using quality inspection equipment. Quality inspection equipment 300 is a device specifically designed for product quality inspection. Applying this welding product quality inspection method to quality inspection equipment enables automated and efficient execution in actual production, improving inspection efficiency and accuracy.

[0033] Exemplarily, the welded product may be a circuit board assembly in a near-eye display device, where the circuit board assembly includes a circuit board body and a battery module welded to the circuit board body.

[0034] The standard image serves as a reference for determining whether welded products are acceptable. It represents the pre-determined acceptable appearance of welded products—the ideal image of a flawless welded product. By comparing the image with the standard image, it is possible to intuitively determine if there are any deviations in the welded product being inspected.

[0035] The Welding Defect Characterization Information Database is a database containing information on various welding defect characteristics. This information includes descriptions of the morphology, location, and distribution of various defect types (such as pores, cracks, lack of fusion, cold welds, continuous welds, and weld line sequences). This database provides an important reference for subsequently determining the actual defect type of welded products.

[0036] For example, the welding defect feature description information library includes: A: cold solder joint; B: solder joint; C: welding line sequence error; D: welding line sequence inversion; E: welding line sequence omission; F: porosity defect; G: crack, etc. The welding defect feature description information can be text information or image information.

[0037] Welded products typically consist of multiple welds, and the quality of each weld affects the overall performance of the product. Including multiple welds in a standard image allows for comprehensive inspection of all welded parts. For example, a standard image may include five welds: weld 1, weld 2, weld 3, weld 4, and weld 5.

[0038] Defect identification, based on experimental data and empirical experience from a large number of welded products, can indicate the likelihood of a specific weld defect at a particular weld joint. For example, certain welds may be more susceptible to porosity defects due to factors such as welding process and material properties. Defect identification for these welds will indicate a higher likelihood of porosity. Defect identification allows for more targeted analysis of potential defects at each weld joint during inspection, improving both accuracy and efficiency.

[0039] It can also be understood that each weld is provided with a defect identifier, which is specifically achieved by collecting a large number of sample images of welded products with different defect types, marking and analyzing the welds in each sample image, and statistically analyzing the probability of each weld having various specific welding type defects. Based on the statistical results, a corresponding defect identifier is set for each weld, and the defect identifier is expressed in the form of a probability value or level.

[0040] For example, a standard image includes solder joints 1, 2, 3, 4, and 5. The defect identifiers for solder joint 1 are: A: Cold solder joint (90%); B: Solder joint (80%). The dotted line identifiers for solder joint 2 are: C: Incorrect soldering sequence (70%); A: Cold solder joint (85%). The defect identifiers for solder joint 3 are: D: Inverted soldering sequence (65%); A: Cold solder joint (75%); B: Solder joint (85%). The defect identifiers for solder joint 4 are: E: Missing soldering sequence (80%); F: Porosity defect (75%). The defect identifiers for solder joint 5 are: G: Crack (85%); A: Cold solder joint (70%).

[0041] In this embodiment of the present application, the quality inspection device 300 first captures an image of the welded product to be inspected and then analyzes and processes the image using the inspection method provided herein. This analysis and processing may include image preprocessing, comparison with a standard image, and defect determination based on defect identification and a weld defect feature description database. By comparing the image of the welded product to be inspected with the standard image, if the two are completely consistent, it means that the welding characteristics of the welded product meet the preset qualification standards and the welded product is judged to be a qualified product; if they are not completely consistent, it means that there may be welding defects and the defect type needs to be further determined.

[0042] After discovering a possible defect in a welded product, the quality inspection device 300 uses the defect identifiers of each weld in the feature image, combined with information from the weld defect feature description database, to determine the likelihood of a specific weld defect type for each inconsistent weld. Through comparison and analysis, the device ultimately determines the actual defect type in the welded product. This helps production personnel promptly identify quality issues in welded products and implement appropriate improvement measures to improve product quality.

[0043] Specifically, refer to Figure 1 , the welding product quality inspection method includes the following steps: S101: Acquire an image to be processed, where the image to be processed is an image of a welding product to be inspected; S102: comparing the image to be processed with the standard image; S103: Determine whether all welding features contained in the image to be processed are completely consistent with those in the standard image; If they are completely consistent, the welding product is determined to be a qualified product; If not completely consistent, the following steps S104-S106 are executed: S104: extracting a feature image based on the image to be processed, wherein the feature image includes a welding spot image that is inconsistent with the welding features in the standard image; S105: Determine the possibility of a specific welding type defect occurring in each inconsistent weld point based on the defect identifier corresponding to each weld point image in the feature image; S106: Based on the possibility of a specific welding type defect occurring in each inconsistent weld point, feature information matching the possibility is retrieved from a welding defect feature description information library, and the feature image is compared with the feature information to determine the actual defect type of the welding product.

[0044] According to an embodiment of the present application, in step S101, an image acquisition device, such as a high-definition camera, equipped on the quality inspection equipment is used to capture an image of the welded product to be inspected. During the acquisition process, it is necessary to ensure that the image is clear and complete, accurately reflecting the appearance characteristics of the welded product, including information such as the shape and location of the welds. In this step, the image to be processed is acquired, providing basic data for subsequent image comparison and analysis. Only by acquiring high-quality images to be processed can the accuracy and reliability of subsequent inspection steps be guaranteed.

[0045] In a specific embodiment, obtaining an image to be processed specifically includes: An industrial camera is used as an image acquisition device. Under specific lighting conditions, the welded product to be inspected is photographed from multiple preset angles. The photographed images are preprocessed to obtain the processed images. The preprocessing includes denoising and contrast enhancement operations.

[0046] In this specific embodiment, the specific lighting conditions are: using a uniformly distributed annular LED light source, and adjusting the brightness of the light source according to the material and surface reflective properties of the welding product to ensure that the collected image is clear and free of shadow interference.

[0047] In this specific embodiment, the welding product to be inspected is photographed from multiple preset angles, including from the front, 45° side, and 90° side, so as to fully obtain the welding feature information of the welding product.

[0048] In step S102, an image comparison algorithm may be used to compare the image to be processed with the standard image pixel by pixel or feature by feature. In step S103, the similarity or difference between the two images may be calculated to determine whether the welding features are completely consistent.

[0049] If the judgment results and treatments are completely consistent, it means that the welding characteristics of the welded product to be inspected meet the preset qualification standards and the welded product is judged to be qualified. At this time, the quality inspection equipment can output a qualified signal to inform the operator that the product has passed the inspection.

[0050] In a specific embodiment, in step S102, comparing the image to be processed with the standard image specifically includes: Extract key welding features from the image to be processed and the standard image; Calculate the similarity between the corresponding features in the image to be processed and the standard image.

[0051] In step S103, it is determined whether all welding features contained in the image to be processed are completely consistent with the standard image, specifically including: When the similarities of all corresponding features exceed a preset threshold, it is determined that all welding features contained in the image to be processed are completely consistent with the standard image.

[0052] In this specific embodiment, the quality comparison and judgment process is mainly divided into two key steps. First, the key welding features in the image to be processed and the standard image are extracted respectively, and then the similarity of the corresponding features is calculated through the feature matching algorithm, and a judgment is made based on the relationship between the similarity and the preset threshold.

[0053] Using an image feature extraction algorithm, key welding features are extracted from both the processed image and the standard image. These key welding features encompass multiple aspects, including the weld's shape (e.g., circular, elliptical, etc.), size (weld diameter or area), location (coordinate position in the image), weld width (weld thickness), and continuity (whether the weld is fractured or discontinuous).

[0054] For example, the image feature extraction algorithm may be an edge detection-based algorithm, which determines the edge of the welding feature by detecting a sudden change in the grayscale value of pixels in the image, and further extracts the key welding feature.

[0055] Image feature extraction extracts representative information from an image for accurate comparison and analysis. Different weld features can reflect the quality of welded products from different perspectives. For example, abnormal weld shape and size may indicate a problem with the welding process, while weld width and continuity directly affect weld strength and sealing. By extracting these key features, it is possible to more accurately determine whether the weld features in the processed image meet standards.

[0056] A feature matching algorithm is used to calculate the similarity between the corresponding features in the processed image and the reference image. Based on the extracted key welding features, the feature matching algorithm establishes a correspondence between the two and then uses a specific calculation method to measure the similarity between the corresponding features. This similarity calculation quantitatively assesses the degree of similarity between the welding features in the processed image and the reference image. A higher similarity indicates that the welding features in the processed image are closer to the reference, and the welded product is more likely to meet quality requirements. Conversely, a lower similarity indicates possible quality issues.

[0057] For example, the feature matching algorithm may be a matching algorithm based on Euclidean distance, which determines the similarity by calculating the Euclidean distance between corresponding feature vectors in the image to be processed and the standard image.

[0058] In step S103, if all welding features contained in the image to be processed are not completely consistent with the standard image, it indicates that the welded product to be inspected may have welding defects, and further subsequent steps need to be performed for detailed analysis to determine the specific defect type.

[0059] In step S104, based on the comparison results of step S103, regions in the image to be processed that are inconsistent with the standard image, i.e., regions with potential defective solder joints, are determined. For example, image segmentation techniques such as threshold segmentation, region growing, and edge detection can be used to extract these inconsistent solder joint images from the image to be processed to form a feature image.

[0060] In this step, the feature image focuses on solder joints that may have defects, reducing the amount of data for subsequent analysis, improving detection efficiency, and enabling more accurate analysis and judgment of defects.

[0061] In a specific implementation, extracting a feature image based on the image to be processed specifically includes: Calculate the difference area between the image to be processed and the standard image, and use the image in the difference area as the preliminary feature image; The preliminary feature image is processed to obtain the final feature image.

[0062] In this embodiment, an image difference algorithm can be used to calculate the difference area between the image to be processed and the standard image. The core principle of the image difference algorithm is to subtract the grayscale values ​​of corresponding pixels in the two images. Specifically, for each pixel position in the image to be processed and the standard image, the grayscale value of the pixel at that position in the image to be processed is subtracted from the grayscale value of the pixel at the corresponding position in the standard image to obtain a difference image. Then, by setting an appropriate threshold, the difference image is binarized, and pixels with a difference greater than the threshold are marked as difference areas. The image formed by these difference areas is used as the preliminary feature image.

[0063] Consider a scenario involving soldering a circuit board. A standard image shows solder joints that are uniformly sized and regular. However, a solder joint in the image being processed has an irregular shape and is larger than normal due to excessive soldering time. Using an image differencing algorithm, we can accurately identify this area that differs from the standard solder joint and use it as part of the preliminary feature image.

[0064] When acquiring the initial feature image, the boundaries of the difference regions may be unclear or small holes may exist due to factors such as image noise and uneven lighting. These small holes can be filled by dilation, making the feature regions in the feature image more continuous and complete, while also enhancing the saliency of the features.

[0065] For example, if there are some tiny cracks or holes in the area representing welding defects in the preliminary feature image, these cracks and holes may be filled through the expansion operation, making the defect area look more complete, which facilitates the subsequent accurate identification and analysis of defects.

[0066] While the dilation operation can fill holes and enhance features, it can also enlarge the feature area and introduce noise that doesn't belong to the feature. The erosion operation can remove this noise introduced by dilation and further refine the feature image boundaries, making the feature image more accurately reflect the actual weld characteristics.

[0067] For example, after the dilation operation, some irregular bumps or burrs may appear around the feature image due to the influence of noise. Through the erosion operation, these bumps and burrs can be removed, making the boundary of the feature image smoother and more accurate.

[0068] After the expansion and corrosion operations, the final feature image with noise interference removed and more complete features is obtained, which provides a reliable basis for subsequent further analysis of welding features and quality judgment.

[0069] In step S105, each weld is assigned a defect indicator in the standard image. These indicators are derived from extensive experimental data and empirical data and reflect the probability of a specific weld defect occurring at that weld. After acquiring the characteristic image, the corresponding weld is located in the standard image based on the positional information of each weld in the characteristic image, and its defect indicator is read. By querying a pre-set defect probability table or using a corresponding calculation model, the probability of a specific weld defect occurring at each inconsistent weld is determined.

[0070] The purpose of this step is to provide an important basis for the subsequent determination of the actual defect type. By analyzing the possibility of different defects in each solder joint, defects can be located and identified more accurately.

[0071] For example, the position information of the weld in the feature image corresponds to the position of weld No. 1 in the standard image. The defect identification of weld No. 1 in the standard image is: A: cold weld, B: continuous weld. The possible welding defects of the welded product are: cold weld, continuous weld, or both cold weld and continuous weld.

[0072] In step S106, the welding defect feature description information library stores feature information of various welding defects, and the feature information can be text information or image information. Preferably, the feature information is image information.

[0073] Text information, such as a detailed description of a defect, including the shape, size range, color characteristics, and location pattern of the defect.

[0074] Image information, i.e., actual photographs or schematic diagrams of welding products with such defects, can intuitively present the appearance of the defects.

[0075] According to the possibility of a specific welding type defect occurring in each inconsistent weld point determined in S106, feature information consistent with the possibility of the specific welding type defect is retrieved from the welding defect feature description information library, and the feature image is compared with the feature information to ultimately determine the actual defect type of the welding product.

[0076] That is to say, in this step, for each inconsistent solder joint, based on the possible defect type obtained from the previous analysis, the feature description of the corresponding defect type is found from the information library, which may be a text description or an image example.

[0077] For example, if the location of a solder joint in the feature image corresponds to the location of solder joint 1 in the standard image, and the defect identification of solder joint 1 in the standard image is: A: cold solder joint, B: continuous solder joint, in this case, the two feature information A: cold solder joint and B: continuous solder joint are found from the information database.

[0078] The acquired feature image is compared with the feature information retrieved from the information library. By comparing the solder joint features in the feature image with the defect features described in the information library, it is determined whether they match. If they match, the actual defect type of the inconsistent solder joint in the welded product can be finally determined. For example, if the feature describing a certain defect in the information library is an irregular protrusion on the solder joint surface, and the solder joint in the feature image also shows a similar irregular protrusion shape, then the actual defect type of the solder joint can be determined to be the defect described in the information library. For another example, if the feature image is consistent with the feature information A: Cold Solder found in the information library, then the actual defect type of the welded product is Cold Solder. If the feature image is consistent with the feature information A: Cold Solder and B: Continuous Solder found in the information library, then the actual defect types of the welded product are Cold Solder and Continuous Solder.

[0079] The collected feature image is compared and analyzed with the corresponding feature information retrieved from the information library. In specific operations, the weld spot features presented in the feature image are compared with the various defect feature descriptions recorded in the information library to determine whether the two are consistent. Under normal circumstances, the collected feature image will match at least one of the welding defect features retrieved from the information library. However, there are special cases where the collected feature image cannot match the welding defect features retrieved from the information library. Once the collected feature image does not match the welding defect features retrieved from the information library, it is necessary to compare each information feature in the welding defect description information library with the collected feature image one by one to determine the defect type corresponding to the feature image, and finally determine the actual defect type of the welding product.

[0080] In a specific example, when the features in the weld defect feature description database are described in text, image processing algorithms are needed to extract the shape characteristics of the inconsistent welds in the feature image, such as outline, aspect ratio, and roundness. For example, for a circular weld, the deviation between its actual outline and the ideal circle can be calculated; for a long weld, the ratio of its length to width can be measured. Furthermore, the texture information of the weld surface in the feature image needs to be analyzed, and the color information of the weld in the feature image needs to be obtained.

[0081] When the information features in the welding defect feature description information database are described in text, it is also necessary to perform natural language processing on the text description to extract key information describing the defect characteristics, such as shape keywords (circular, oval, irregular, etc.), texture keywords (rough, smooth, granular, etc.), and color keywords (red, black, metallic luster, etc.).

[0082] The extracted image features are compared with the key features obtained from the parsed text description. For shape features, the similarity of contours can be calculated, such as using the Hausdorff distance to measure the similarity between two contours. For texture features, the Euclidean distance or cosine similarity of texture statistics can be calculated. For color features, the intersection of color histograms or the Bhattacharyya distance can be calculated.

[0083] The purpose of this step is to accurately identify the actual defect type of the welded product, providing a clear direction for subsequent quality improvement and product repair. At the same time, the test results are recorded in the database of the quality inspection equipment to facilitate statistical analysis and traceability of welding quality.

[0084] The welding product quality inspection method provided in the embodiments of the present application is applied to quality inspection equipment, which pre-stores a standard image and a welding defect feature description information library. The standard image corresponds to a preset qualified form of the welding product, and the standard image includes multiple welds, each of which is provided with a defect identifier to indicate the likelihood of a specific welding type defect occurring at that weld. By comparing the collected image of the welding product to be inspected with the standard image, extracting the feature image, and determining the actual defect type of the welding product based on the defect identifier of the weld and the welding defect feature description information library, the quality of the welding product can be more accurately and comprehensively inspected, improving the accuracy and reliability of the inspection while also increasing the inspection efficiency.

[0085] According to an embodiment of the present application, an image acquisition device is used to acquire an image of a welding product to be inspected, and before acquiring the image to be processed, the following steps are further included: Acquire employee hand motion images, analyze and process the acquired employee hand motion images based on a pre-built motion analysis model, and determine whether the conditions for using an image acquisition device to acquire images of the welding product to be inspected are currently met based on the analysis results. If so, perform subsequent acquisition operations; if not, remind the employee to perform hand motion operations.

[0086] In this embodiment, before officially starting to capture images of the welded product to be inspected, the system first captures images of the employee's hand movements. For example, the image capture module captures images of the employee's hand movements. The employee's hand movements often reflect the relevant processing operations performed on the welded product to be inspected.

[0087] Specifically, obtaining images of employees' hand movements can cover the following situations: The first is to capture images of employees wearing wristbands designed to prevent electrical breakdown in welding products. Capturing images of employees wearing these wristbands indicates effective electrical isolation between the employee and the welded product being inspected, preventing electrical breakdown and ensuring safety during the operation.

[0088] The second is to capture images of workers wiping the surface of welded parts. Capturing the hand movements of workers wiping welded parts indicates that impurities such as welding slag have been removed from the surface of the welded parts, facilitating accurate image acquisition and quality inspection.

[0089] The third step is to capture an image of the worker wiping the welded product surface and placing it back in its original area. Capturing this image indicates that the welded product to be inspected has been properly returned to its original area, meeting the requirements for subsequent image acquisition.

[0090] For example, the above three employee hand motion images need to be acquired in sequence.

[0091] For example, the first step is to capture an image of an employee wearing a wristband that prevents electrical breakdown on a welding product. This step captures an image of the employee wearing the wristband, which means that the employee has been effectively electrically isolated from the welding product to be inspected.

[0092] The second step is to capture an image of the worker wiping the welded part. After completing the electrical isolation operation, the worker needs to clean the surface of the welded part to be inspected. Capturing the worker's hand motion while wiping the welded part indicates that impurities such as welding slag have been removed from the surface.

[0093] The third step is to capture an image of the welded part after the worker has wiped it and returned it to its original location. After cleaning the surface, the worker must properly place the welded part to be inspected in its designated location. Capturing this image indicates that the welded part has been returned to its original location, meeting the requirements for subsequent image acquisition. Only when the part is in the correct position can the image capture device accurately capture it according to the preset parameters and angles, generating high-quality images for processing and providing reliable evidence for subsequent quality inspections.

[0094] Furthermore, capturing images of the employee's hand movements can also include detecting whether the employee's hand has crossed the centerline of the workbench. If such images are captured, it indicates that the employee may have violated established safety operating procedures or may have manipulated the welding product beyond the specified range. If images of the employee's hand crossing the centerline of the workbench are captured, the employee should be promptly alerted to avoid potential risks and adverse consequences.

[0095] In this embodiment, to ensure the effective use of the acquired image information, it is necessary to conduct in-depth analysis and processing of these images with the help of a pre-built motion analysis model. This motion analysis model is obtained through extensive sample training and optimization, and has the ability to accurately identify hand movement characteristics and states.

[0096] Based on the results of this model analysis, it is determined whether the conditions for capturing images of the welded product to be inspected using an image acquisition device are currently met. This judgment criteria is based on a comprehensive set of factors, including actual production needs, safety regulations, and quality inspection requirements. If the analysis results indicate that the current conditions are met, the system will quickly and accurately execute subsequent image acquisition operations to ensure that high-quality images of the welded product to be inspected are acquired in a timely manner, providing a reliable basis for subsequent quality inspections. Conversely, if the analysis results indicate that the current conditions are not met, the system will immediately trigger a reminder mechanism, clearly and unambiguously reminding employees to perform the appropriate hand movements until the acquisition conditions are met, thus ensuring the standardization and accuracy of the entire image acquisition process.

[0097] According to an embodiment of the present application, after determining the actual defect type of the welding product, the method further includes: The inspection results are recorded and stored, including the welding product number, inspection time, defect type and defect location information; and the inspection results are fed back to the production control system.

[0098] In this embodiment, comprehensive and accurate recording and storage of inspection results are performed. Specifically, this includes recording the unique number of the welded product to enable precise traceability of each product; recording the inspection time with appropriate accuracy to provide a basis for subsequent analysis of production process timing; recording the defect type to clearly define the specific defect category present in the product; and recording the defect location information to accurately describe the specific location of the defect on the product, which can be achieved through coordinate positioning or other appropriate positioning methods.

[0099] Through such comprehensive and detailed recording and storage, a solid and reliable data foundation is provided for subsequent data analysis, quality traceability and process improvement.

[0100] At the same time, test results are promptly fed back to the production control system, allowing it to quickly make adjustments and decisions based on the test results. For example, based on the defect type and location, production equipment parameters can be optimized and production processes adjusted to prevent similar defects from recurring. Alternatively, defective products can be classified and processed, with subsequent actions such as rework and scrapping arranged. This ensures efficient and stable production and improves overall product quality.

[0101] <Device Example> According to an embodiment of the present application, a welding product quality detection device 200 is also provided. The detection device is applied to a quality inspection device, wherein the quality inspection device pre-stores a standard image and a welding defect feature description information library; wherein, The standard image corresponds to a preset qualified form of the welded product, and the standard image includes a plurality of welds, each of which is provided with a defect identifier, wherein the defect identifier is used to indicate the possibility of a specific welding type defect occurring in the weld; Reference Figure 2 , the welding product quality detection device 200 includes: An image acquisition module 201 is used to acquire an image to be processed, where the image to be processed is an image of a welding product to be inspected; A quality detection module 202 is configured to compare the image to be processed with the standard image; A judgment module 203 judges whether all welding features contained in the image to be processed are completely consistent with the standard image; If they are completely consistent, the welding product is determined to be a qualified product; If they are not completely consistent, the extraction module 204 extracts a feature image from the image to be processed, wherein the feature image includes an image of a welding spot that is inconsistent with the welding feature in the standard image; A determination module 205 determines the possibility of a specific welding type defect occurring in each inconsistent weld point based on the defect identifier corresponding to each weld point image in the feature image; The comparison module 206 retrieves feature information that matches the possibility of a specific welding type defect occurring in each inconsistent weld point from a welding defect feature description information library, compares the feature image with the feature information, and determines the actual defect type of the welding product.

[0102] In an embodiment of the present application, a welding product quality inspection device is applied to a quality inspection device, which pre-stores a standard image and a welding defect feature description information library. The standard image corresponds to a preset qualified form of the welding product, and the standard image includes multiple welds, each of which is provided with a defect identifier to indicate the likelihood of a specific welding type defect at that weld. By comparing the collected image of the welding product to be inspected with the standard image, extracting the feature image, and determining the actual defect type of the welding product based on the defect identifier of the weld and the welding defect feature description information library, the quality of the welding product can be more accurately and comprehensively inspected, improving the accuracy and reliability of the inspection, while also increasing the inspection efficiency.

[0103] It should be noted that the specific implementation of the welding product quality inspection device of the embodiment of the present application can refer to the various embodiments of the welding product quality inspection method mentioned above, and therefore at least has all the beneficial effects brought by the technical solutions of the above embodiments, which will not be repeated here one by one.

[0104] An embodiment of the present application also provides a quality inspection device, which includes the welding product quality inspection device 200 as described in the second aspect; or includes a processor and a memory, the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute the welding product quality inspection method as described in the first aspect.

[0105] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements any one of the welding product quality detection methods provided in the above method embodiments.

[0106] The present application may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present application.

[0107] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or raised-in-groove structure on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0108] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0109] The computer program instructions for performing the operations of the present application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, the state information of the computer-readable program instructions is used to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), so that the electronic circuit can execute the computer-readable program instructions, thereby implementing various aspects of the present application.

[0110] Various aspects of the present application are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0111] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0112] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0113] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

[0114] The embodiments of the present application have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, their practical applications, or technical improvements in the marketplace, or to enable other persons skilled in the art to understand the embodiments disclosed herein. The scope of this application is defined by the appended claims.

Claims

1. A welding product quality detection method, characterized in that: The method is applied to a quality inspection device, wherein the quality inspection device pre-stores a standard image and a welding defect feature description information library; wherein, The standard image corresponds to a preset qualified form of the welded product, and the standard image includes a plurality of welds, each weld being provided with a defect identifier, wherein the defect identifier is used to indicate the possibility of a specific welding type defect occurring in the weld; The welding product quality detection method comprises the following steps: Acquire an image to be processed, where the image to be processed is an image of a welding product to be inspected; Comparing the image to be processed with the standard image; Determine whether all welding features contained in the image to be processed are completely consistent with the standard image; if they are completely consistent, determine that the welded product is a qualified product; if they are not completely consistent, extract a feature image based on the image to be processed, wherein the feature image includes images of weld spots that are inconsistent with the welding features in the standard image; Determining the likelihood of a specific welding type defect occurring in each inconsistent weld point based on the defect identifier corresponding to each weld point image in the feature image; Based on the possibility of a specific welding type defect occurring in each inconsistent weld spot, feature information matching the possibility is retrieved from a welding defect feature description information library, and the feature image is compared with the feature information to determine the actual defect type of the welded product.

2. The welding product quality detection method according to claim 1, characterized in that: The comparing the image to be processed with the standard image specifically includes: Extract key welding features from the image to be processed and the standard image; Calculate the similarity between the corresponding features in the image to be processed and the standard image.

3. The welding product quality detection method according to claim 1, characterized in that: The step of extracting a feature image based on the image to be processed specifically includes: Calculate the difference area between the image to be processed and the standard image, and use the image in the difference area as the preliminary feature image; The preliminary feature image is processed to obtain the final feature image.

4. The welding product quality detection method according to claim 1, characterized in that: Each solder joint is provided with a defect mark, specifically including: Collect sample images of welding products with different defect types, annotate and analyze the welds in each sample image, calculate the probability of each weld having various specific welding defects, and set a corresponding defect identifier for each weld based on the statistical results. The defect identifier is expressed as a probability value or level.

5. The welding product quality detection method according to claim 1, characterized in that: The obtaining of the image to be processed specifically includes: The welded product to be inspected is photographed from multiple preset angles, and the photographed images are preprocessed to obtain images to be processed, wherein the preprocessing includes denoising and contrast enhancement operations.

6. The welding product quality detection method according to claim 1, characterized in that: Before obtaining the image to be processed, it also includes: Acquire employee hand motion images, analyze and process the acquired employee hand motion images based on a pre-built motion analysis model, and determine whether the conditions for collecting images of the welding product to be inspected are met based on the analysis results. If so, perform subsequent collection operations. If not, remind the employee to perform hand motion operations.

7. The welding product quality detection method according to claim 6, characterized in that: Acquiring employee hand motion images specifically includes: Acquire an image of an employee wearing a wristband to prevent electrical breakdown of a welding product, acquire an image of an employee wiping the surface of a welding product, and acquire an image of an employee wiping the surface of a welding product and placing the welding product in the original area.

8. The welding product quality detection method according to claim 1, characterized in that: After determining the actual defect type of the welding product, it also includes: The inspection results are recorded and stored, including the welding product number, inspection time, defect type and defect location information; and the inspection results are fed back to the production control system.

9. A welding product quality detection device, characterized in that: Applied to quality inspection equipment, the quality inspection equipment pre-stores standard images and welding defect feature description information library; wherein, The standard image corresponds to a preset qualified form of the welded product, and the standard image includes a plurality of welds, each of which is provided with a defect identifier, wherein the defect identifier is used to indicate the possibility of a specific welding type defect occurring in the weld; The welding product quality detection device comprises: An image acquisition module is used to acquire an image to be processed, wherein the image to be processed is an image of the welded product to be inspected; A quality detection module, configured to compare the image to be processed with the standard image; a judgment module for judging whether all welding features contained in the image to be processed are completely consistent with those in the standard image; if they are completely consistent, the welded product is determined to be a qualified product; if they are not completely consistent, a feature image is extracted from the image to be processed by an extraction module, the feature image including images of weld spots that are inconsistent with the welding features in the standard image; a determination module, determining the possibility of a specific welding type defect occurring in each inconsistent weld point based on the defect identifier corresponding to each weld point image in the feature image; The comparison module retrieves feature information that matches the possibility of a specific welding type defect occurring in each inconsistent weld point from a welding defect feature description information library, compares the feature image with the feature information, and determines the actual defect type of the welded product.

10. A quality inspection device, characterized in that: The quality inspection equipment includes the welding product quality inspection device according to claim 9; or, the quality inspection equipment includes a processor and a memory, the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute the welding product quality inspection method according to any one of claims 1 to 8.

Images