Differential case defect detection method and device, computer device and storage medium

Abstract

The application relates to the technical field of automobile accessory detection, and relates to a differential shell defect detection method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: positioning a differential shell by using an installation positioning module of a detection station, starting a light source module and a code scanning module, and identifying and recording the number information of the differential shell; collecting external images of the differential shell at different angles; performing defect detection on the external images at different angles based on a preset defect identification dataset, and outputting a defect detection result; and adjusting the preset defect identification dataset according to the defect detection result. The method can improve the detection efficiency and accuracy of the differential shell and ensure product quality.

Description

Technical Field

[0001] This application relates to the field of automotive parts testing technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for detecting defects in differential housings. Background Technology

[0002] The new energy vehicle industry has developed rapidly under the guidance of national policies and driven by technological innovation. New energy transmission products generally include hybrid transmissions and reducers, with the differential being a crucial component. It transmits torque and enables differential speeds between the left and right vehicles. A differential assembly typically consists of a differential housing, planetary gears, half-shaft gears, and planetary gear shafts. The differential housing, as a major component, primarily transmits torque and supports the normal operation of the internal bevel gears and main reduction gears. Problems with the quality of the differential housing can lead to abnormal operation or even breakage and failure, posing a reliability risk.

[0003] Traditionally, visual inspection relies on manual labor, which is prone to subjectivity and inaccurate assessments, hindering quality control. Therefore, there is an urgent need for a method, device, computer equipment, computer-readable storage medium, and computer program product for detecting defects in differential housings, capable of improving the efficiency and accuracy of differential housing inspections and ensuring product quality. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for detecting differential housing defects, which can improve the efficiency and accuracy of differential housing inspection and ensure product quality, in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a method for detecting defects in a differential housing, including:

[0006] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0007] Acquire external images of the differential housing from different angles;

[0008] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0009] Adjust the preset defect identification dataset based on the defect detection results.

[0010] In one embodiment, before positioning the differential housing using the installation positioning module at the testing station, the method further includes:

[0011] Collect external images of at least one standard defect-free sample to construct the preset defect identification dataset, which includes standard comparison images and a defect learning set;

[0012] The camera module is calibrated by using a ruler of a fixed size to perform data calibration and determine the ratio of image pixels to their actual size.

[0013] In one embodiment, after outputting the defect detection result, the method further includes:

[0014] If the defect detection results meet the requirements, the defect detection results and the corresponding number information are stored.

[0015] If the defect detection results do not meet the requirements, an error value will be reported and an alarm will be triggered.

[0016] In one embodiment, acquiring external images of the differential housing from different angles includes:

[0017] The differential housing is divided into key and non-key areas. The key areas include the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window.

[0018] External images from different angles were collected for both key areas and other areas, with the image collection frequency for key areas being higher than that for other areas.

[0019] In one embodiment, the detection station integrates a positioning module, a barcode scanning module, and a light source module;

[0020] The installation and positioning module includes a clamping unit, which is used to assemble the differential housing onto the testing station at a fixed position and a fixed angle.

[0021] In one embodiment, the step of performing defect detection on external images from different angles based on a preset defect recognition dataset and outputting defect detection results includes:

[0022] Acquire image features from external images at different angles, and convert the image features from image format to data format;

[0023] Based on image features converted to data format and a pre-set defect recognition dataset, the defect type and defect size are identified.

[0024] Output the defect detection results based on the defect type and defect size.

[0025] Secondly, this application also provides a differential housing defect detection device, comprising:

[0026] The device includes a testing station, a defect detection module, and a data update module. The testing station includes an installation and positioning module, an image acquisition module, a barcode scanning module, and a light source module.

[0027] The installation positioning module is used to position the differential housing; the light source module and the barcode scanning module are used to identify and record the serial number information of the differential housing.

[0028] The image acquisition module is used to acquire external images of the differential housing from different angles;

[0029] The defect detection module is used to perform defect detection on external images from different angles in real time based on a preset defect recognition dataset and output the defect detection results.

[0030] The data update module is used to adjust the preset defect identification dataset based on the defect detection results.

[0031] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0032] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0033] Acquire external images of the differential housing from different angles;

[0034] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0035] Adjust the preset defect identification dataset based on the defect detection results.

[0036] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0037] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0038] Acquire external images of the differential housing from different angles;

[0039] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0040] Adjust the preset defect identification dataset based on the defect detection results.

[0041] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0042] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0043] Acquire external images of the differential housing from different angles;

[0044] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0045] Adjust the preset defect identification dataset based on the defect detection results.

[0046] The aforementioned differential housing defect detection method, apparatus, computer equipment, computer-readable storage medium, and computer program products, through automated inspection processes, reduce the subjectivity and inaccuracy of manual inspection, thereby improving the efficiency and accuracy of differential housing defect detection. Utilizing machine vision inspection technology, automated inspection can be achieved on the production line, significantly reducing labor costs and improving production efficiency. Through precise image processing and defect recognition, surface and internal defects of the differential housing, such as porosity, flow marks, cracks, and scratches, can be effectively detected, thereby improving product quality and reliability. Real-time inspection of the differential housing and rapid output of inspection results are possible, helping to promptly identify and address quality issues and reduce the risk of defective products entering the market. Feedback from actual inspection results allows for continuous adjustment and optimization of the preset defect identification dataset, making the defect detection model more accurate and improving detection accuracy. Precise defect detection can effectively identify non-conforming products, reduce scrap rates, and improve material utilization and production efficiency. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is an application environment diagram of a differential housing defect detection method in one embodiment;

[0049] Figure 2 This is a flowchart illustrating a differential housing defect detection method in one embodiment;

[0050] Figure 3 This is a flowchart illustrating a differential housing defect detection method in another embodiment;

[0051] Figure 4 This is a structural block diagram of the differential housing defect detection device in the most detailed embodiment;

[0052] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0054] The differential housing defect detection method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0055] Terminal 102 is used to position the differential housing using the installation and positioning module of the inspection station, activate the light source module and the barcode scanning module, identify and record the serial number information of the differential housing; and collect external images of the differential housing from different angles. Server 104 is used to perform defect detection on the external images from different angles based on a preset defect identification dataset, output the defect detection results, and adjust the preset defect identification dataset according to the defect detection results.

[0056] The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle systems, and projection devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0057] In one exemplary embodiment, such as Figure 2 As shown, a method for detecting defects in a differential housing is provided, which can be applied to... Figure 1Taking the terminal in the example, the explanation includes the following steps S202 to S208. Wherein:

[0058] Step S202: Use the installation and positioning module of the inspection station to position the differential housing, start the light source module and the barcode scanning module to identify and record the serial number information of the differential housing.

[0059] Specifically, the safety inspection station includes a positioning module. This device ensures that the differential housing is placed in a fixed position and angle on the inspection station. This guarantees that the position and angle of the differential housing are consistent during each inspection, thus ensuring the consistency and accuracy of the inspection.

[0060] The light source module provides uniform and sufficient illumination so that the camera module can clearly capture images of the differential housing from various angles. The barcode scanning module is used to identify the QR code or other identifiers on the differential housing. This QR code contains unique identification information for the differential housing, such as the product number and production batch. By scanning the code, the system can record and track the inspection information of each differential housing, which is crucial for subsequent quality control and product traceability.

[0061] Once the differential housing is correctly placed and clamped, the sensor recognizes that the housing is clamped and sends a signal to the industrial control computer, indicating that inspection can begin. Simultaneously, the barcode scanning module reads the QR code information on the differential housing and records it in the system. This allows the inspection result of each differential housing to be associated with its serial number, facilitating subsequent data analysis and quality tracking.

[0062] Step S204: Acquire external images of the differential housing from different angles.

[0063] Specifically, the differential housing is a three-dimensional structure with multiple surfaces and angles. To ensure comprehensive inspection, images need to be acquired from different angles to cover all important areas of the differential housing. Multiple cameras are arranged around the differential housing using a camera module, which can simultaneously or sequentially capture images of the housing from various angles. This ensures that every detail of the housing is captured from different perspectives. The acquired images need sufficient resolution and sharpness so that subsequent image processing and defect detection algorithms can accurately identify defects.

[0064] To ensure image quality, the light source module provides uniform illumination, eliminating shadows and reflections, making defects in the image more noticeable.

[0065] The acquired images are fed into the image processing module of the industrial control computer for feature extraction and defect identification. This may involve using image processing algorithms (such as SIFT, SURF, ORB, etc.) to identify and mark potential defects. Data calibration is performed using a fixed-size scale to determine the actual size represented by individual pixels in the image, thus enabling the calculation of the actual area of ​​the defect. Perspective transformation is then performed to correct the image and obtain accurate data.

[0066] Step S206: Based on the preset defect recognition dataset, perform defect detection on external images from different angles and output the defect detection results.

[0067] Specifically, a pre-defined defect identification dataset containing various known defect types is constructed. This dataset typically includes images of differential housings in normal and various defective states, along with detailed annotation information for these images, such as defect location, type, and size.

[0068] The acquired external images undergo preprocessing, including noise reduction, contrast enhancement, brightness adjustment, and color balance adjustment, to improve image quality and make defect features more prominent. Image processing techniques are used to extract key feature points and descriptors from the images; these feature points and descriptors represent important information in the image, such as edges, corners, and textures. The extracted features are compared with features in a pre-defined defect recognition dataset, and machine learning algorithms are used to identify defects in the images. This may involve using a classifier to determine the presence and type of defects in the image.

[0069] Once a defect is identified, the system needs to pinpoint its precise location and classify it into a specific type from a pre-defined dataset. For each identified defect, the system needs to calculate its size and area. This is typically done by converting pixel dimensions to actual dimensions, potentially involving perspective conversion and ruler calibration. Based on pre-defined rules and standards, the detected defect is evaluated to determine whether the differential housing should be accepted or rejected.

[0070] The system outputs the defect detection results, which may include the location, type, size, area of ​​the defect, and whether it is acceptable. If the detection results do not meet the requirements, the system will issue an alarm through the feedback module and stop the detection process, awaiting manual intervention. Acceptable detection results will be stored and linked to the differential housing's serial number for subsequent quality tracking and traceability.

[0071] Step S208: Adjust the preset defect identification dataset based on the defect detection results.

[0072] Specifically, results are collected from actual defect detection processes, including correctly identified defects and false positives or false negatives. For false positives (the system incorrectly identifies a non-defect as a defect) and false negatives (the system fails to identify an actual defect), detailed analysis is performed to determine the cause of the problem. Images of false positives and false negatives are added to a pre-defined defect identification dataset as new training samples. This helps the system learn and identify defect types that were previously not correctly identified.

[0073] Retrain the machine learning model using the updated dataset, including supervised learning algorithms such as convolutional neural networks (CNNs) or other image recognition models. Adjust and optimize algorithm parameters based on the new dataset and detection results to improve defect identification accuracy. After updating the dataset and model, perform validation and testing to ensure the improved system can identify defects more accurately. Continuously monitor the performance of the defect detection system to ensure it remains efficient and accurate in real-world production environments. Establish a feedback loop so that each detection result can be used to further optimize and adjust the pre-set defect identification dataset.

[0074] The aforementioned differential housing defect detection method reduces the subjectivity and inaccuracy of manual inspection through an automated inspection process, improving the efficiency and accuracy of differential housing defect detection. Utilizing machine vision inspection technology enables automated inspection on the production line, significantly reducing labor costs and increasing production efficiency. Through precise image processing and defect recognition, surface and internal defects of the differential housing, such as porosity, flow marks, cracks, and scratches, can be effectively detected, thereby improving product quality and reliability. Real-time inspection of the differential housing and rapid output of results facilitate timely detection and handling of quality issues, reducing the risk of defective products entering the market. Feedback from actual inspection results allows for continuous adjustment and optimization of the preset defect recognition dataset, making the defect detection model more accurate and improving detection accuracy. Precise defect detection effectively identifies non-conforming products, reducing scrap rates and improving material utilization and production efficiency.

[0075] In one exemplary embodiment, such as Figure 3 As shown, before positioning the differential housing using the installation and positioning module at the testing station, the following steps are also included:

[0076] Step S302: Collect external images of at least one standard defect-free sample and construct a preset defect identification dataset, which includes standard comparison images and a defect learning set;

[0077] Step S304: The camera module is calibrated by using a ruler of fixed size to perform data calibration and determine the ratio of image pixels to actual size.

[0078] Specifically, the purpose of acquiring external images of standard defect-free samples is to establish a high-quality benchmark dataset for subsequent defect detection and comparison. At least one known defect-free differential housing sample is selected, and its external images are acquired from multiple angles using a camera module. These images are stored as "standard defect-free images" in a pre-defined defect identification dataset for comparison with images of actually detected differential housings.

[0079] Next, a pre-defined defect identification dataset is constructed. The standard comparison images include standard defect-free sample images collected from multiple angles, which will serve as the benchmark for detection. The defect learning set includes images of various known defects, such as shrinkage porosity, pinholes, dents, and cracks. These images are used to train a machine learning model to identify and classify these defects.

[0080] The purpose of calibrating the camera module is to ensure that the images captured by the camera module are consistent with the dimensions of the actual objects, so as to facilitate accurate measurement and identification of defects. Operation: Data calibration is performed using a fixed-size ruler (such as a calibration plate or calibration block). The ruler usually has patterns or markings of known dimensions used to determine the proportional relationship between pixels in the image and their actual dimensions. Data calibration: By comparing the pixel dimensions of the ruler in the image with their actual physical dimensions, the actual length represented by each pixel is calculated. This helps to convert the measurements in the image into actual physical dimensions, thereby accurately identifying and measuring defects.

[0081] The calibration process determines the actual length corresponding to each pixel in the image (e.g., each pixel represents 0.1 mm). In subsequent defect detection, this ratio is used to convert the defect size in the image into the actual physical size for accurate defect analysis and judgment.

[0082] In this embodiment, the above steps ensure that the differential housing defect detection system is calibrated and ready before actual testing begins, thereby improving the accuracy and reliability of the testing.

[0083] In one exemplary embodiment, after outputting the defect detection results, the method further includes:

[0084] If the defect detection results meet the requirements, the defect detection results and corresponding number information will be stored.

[0085] If the defect detection results do not meet the requirements, an error value will be reported and an alarm will be triggered.

[0086] Specifically, when defect inspection results show that the differential housing meets preset quality standards and requirements, the inspection results and related serial number information are stored in a database or data storage system. This typically includes detailed information such as the location, type, size, and area of ​​the defect, as well as the differential housing serial number. Ensuring that the inspection results for each differential housing are correctly associated with its serial number facilitates subsequent quality tracking and traceability. These records can be used for quality control reports, statistical analysis, and historical data review to monitor the stability and improvement of the production process.

[0087] When defect inspection results show that the differential housing does not meet preset quality standards and requirements, the system automatically marks the non-compliant inspection result as an error and feeds it back to the operator or control system. It triggers alarm systems, such as audible and visual alarms or automatic message notifications, to immediately notify relevant personnel to handle the non-conforming differential housing. In some cases, the system may automatically stop the inspection process until the problem is resolved to prevent non-conforming products from entering the next production stage. All non-compliant inspection results are recorded, including error type, time, and the relevant differential housing number, to facilitate subsequent problem analysis and corrective actions. For non-conforming differential housings, further processing is required according to the factory's non-conforming product handling procedures, which may include rework, scrapping, or isolation. Using the stored inspection results and problem records, the root cause of the quality problem is analyzed, and corresponding quality improvement measures are taken.

[0088] In this embodiment, these steps ensure that the defect detection of the differential housing is not only accurate and efficient, but also that quality issues can be addressed in a timely manner, thereby improving overall production quality and efficiency.

[0089] In one exemplary embodiment, acquiring external images of the differential housing from different angles includes:

[0090] Divide the differential housing into key areas and non-key areas. Key areas include the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window.

[0091] External images from different angles were collected for both key areas and other areas, with the image collection frequency for key areas being higher than that for other areas.

[0092] Specifically, the purpose of dividing the area into key areas and non-key areas is to improve the efficiency and accuracy of the inspection. First, it is necessary to identify the key areas on the differential housing, which have a significant impact on the function and performance, and therefore require more detailed inspection.

[0093] Key areas include, but are not limited to, the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window. These areas require special attention due to their functional importance.

[0094] Non-key areas: Other parts besides key areas also need to be inspected, but at a lower frequency.

[0095] Acquire external images of key areas:

[0096] High-frequency acquisition: Due to the importance of key areas, images need to be acquired at a higher frequency to ensure that any possible defects can be captured.

[0097] Multi-angle coverage: Ensure that images of key areas are acquired from multiple angles to comprehensively check the external quality of these areas.

[0098] Image quality: For images of key areas, higher resolution and clearer image quality may be required to facilitate subsequent defect detection.

[0099] Acquire external images of non-critical areas:

[0100] Lower frequency acquisition: In non-critical areas, a lower image acquisition frequency can be used to balance the detection coverage and efficiency.

[0101] Basic coverage: Although the frequency is low, it is still necessary to ensure basic image coverage of non-key areas from different angles in order to detect obvious defects.

[0102] Specific operations for image acquisition:

[0103] Camera module layout: Based on the shape and size of the differential housing, the camera modules are arranged reasonably to ensure coverage of all key and non-key areas.

[0104] Light source control: Adjust the light source to ensure sufficient illumination at different angles, especially in image acquisition of key areas.

[0105] Automated control: The camera module's shooting frequency and angle, as well as the adjustment of the light source, are controlled through an automated control system.

[0106] Finally, the image acquisition frequency and angle for each region, along with the corresponding image data, are recorded. The acquired images are preprocessed, such as denoising and contrast enhancement, to improve the accuracy of subsequent defect detection.

[0107] In this embodiment, a targeted image acquisition strategy can ensure that critical areas of the differential housing receive sufficient attention, while maintaining the efficiency and cost-effectiveness of the overall inspection process. This helps to improve the quality control level of the differential housing and reduce potential failure risks.

[0108] In one exemplary embodiment, the inspection station integrates a positioning module, a barcode scanning module, and a light source module.

[0109] The mounting and positioning module includes a clamping unit, which is used to assemble the differential housing into the testing station at a fixed position and angle.

[0110] Specifically, the function of the mounting positioning module is to ensure that the differential housing maintains the correct position and angle during testing. The clamping unit, part of the mounting positioning module, is used to secure the differential housing, ensuring it does not move or rotate during testing. Once the differential housing is placed on the testing station, the clamping unit is automatically or manually activated to assemble the differential housing in a fixed position and angle. The clamping unit needs to be sufficiently robust to withstand the forces that may be generated during testing, while ensuring that it does not damage the differential housing.

[0111] In this embodiment, by positioning the differential housing, it can be ensured that it will not move or rotate during the detection process, which is beneficial for acquiring more accurate external images.

[0112] In one exemplary embodiment, based on a preset defect recognition dataset, defect detection is performed on external images from different angles, and the defect detection results are output, including:

[0113] Acquire image features from external images at different angles, and convert the image features from image format to data format;

[0114] Based on image features converted to data format and a pre-set defect recognition dataset, the defect type and defect size are identified.

[0115] Output the defect detection results based on the defect type and defect size.

[0116] Specifically, a camera module is used to acquire external images of the differential housing from different angles. Image processing techniques (such as edge detection, corner detection, and texture analysis) are then used to extract key features from the images. Commonly used feature extraction algorithms include SIFT, SURF, and ORB. The extracted image features are then converted from image format to data format, typically numerical data, to facilitate computer processing and analysis. This includes converting pixel values, color information, and texture information into vector or matrix forms.

[0117] The transformed data features are compared and analyzed with features in a pre-set defect identification dataset. Machine learning or deep learning models (such as support vector machines, neural networks, etc.) are used to identify defect types in the images based on the pre-set defect identification dataset. For identified defects, their dimensions are measured using image processing techniques (such as pixel counting, geometric analysis, etc.), and the pixel dimensions are converted into actual physical dimensions.

[0118] Based on the identified defect types and sizes, as well as the preset quality standards, determine whether the differential housing meets the quality requirements. This includes determining the specific location of the defect on the differential housing. The types of identified defects, such as cracks, depressions, protrusions, etc. The actual physical sizes of the defects, including length, width, depth, etc. According to the preset standards, determine whether the differential housing is qualified. Record the test results in the database and provide feedback to the production process as needed, such as alarm, stop the production line, etc.

[0119] In this embodiment, through the above steps, automated and intelligent defect detection is achieved, improving the detection efficiency and accuracy, reducing the interference of human factors, and ensuring the consistency and reliability of product quality.

[0120] One of the most detailed embodiments of this application is:

[0121] Such as Figure 4 As shown, the differential housing defect detection device based on vision detection includes a detection station, a camera module, and an industrial control computer.

[0122] The detection station is used to place the differential housing. The detection station should include an installation positioning device, sensors, a barcode scanning module, and a light source module. The installation positioning module needs to ensure that the differential housing can be assembled onto the detection station at a fixed position and a fixed angle, and should include a reliable clamping module; the sensors can include weight sensors or displacement sensors, which are used to identify whether the differential housing has been clamped and transmit it to the industrial control computer to identify whether to start detection or whether it has been taken away after detection; the barcode scanning module is used to identify the QR code on the differential housing and record the information of the currently detected differential housing; the light source module is used to irradiate the differential housing to facilitate the identification of images at various angles.

[0123] The camera module includes multiple cameras, which are used to capture images of the differential housing at various angles.

[0124] The industrial control computer is used to obtain image data, learn data, process data, and determine whether the detection meets the requirements. The industrial control computer should include: a rule compilation module, an image processing module, a self-learning module, a data analysis module, a storage module, and a feedback module. The rule compilation module is used to input the off-line detection determination requirements; the image processing module is used to obtain the images in the camera module, record the standard defect-free images and the detection images, and process the images into data; the self-learning module is used to store the results in the feature library when defects are detected and improve the defect identification standards; the data analysis module compares the standard defect-free images, identifies the defect categories and calculates the defect sizes of the detected parts, and determines the data based on the rule requirements; the storage module is used to record the current calculation results and images when the test results meet the determination requirements; the feedback module is used to alarm and stop the detection when the test results do not meet the determination requirements.

[0125] Visual inspection-based methods for detecting defects in differential housings include:

[0126] Before testing, a standard comparison chart and a defect learning set should be constructed. Information on multiple standard defect-free samples should be entered into the same testing station as the testing standard. At the same time, various defective samples should be entered into the testing station, and defect areas should be delineated and defect categories marked, such as shrinkage porosity, sand holes, dents and cracks, so that the machine can learn automatically and be used for subsequent defect type identification.

[0127] The camera module is calibrated using a fixed-size scale to determine the actual size represented by each pixel in the image. The defect area can be obtained by calculating the number of pixels. Perspective transformation and image correction are also necessary to obtain accurate data.

[0128] Specialized defect samples must be inspected before and after the production line starts to ensure that the inspection equipment can effectively identify and accurately determine the type, area, and size of defects. The formal inspection process is as follows:

[0129] 1.1 Place the differential housing on the installation and positioning module at the testing station. When the sensor recognizes the correct sample information, it will automatically clamp and start the corresponding testing process.

[0130] 1.2 After the testing process begins, the light source and barcode scanner are activated to record the current assembly number information;

[0131] 1.3 The camera module is arranged around the differential housing. When shooting is started, the camera module takes pictures according to the detection requirements. The arrangement requirements should be able to completely capture the differential housing, so that image acquisition can be completed without rotating the detection table, thereby improving detection efficiency and accuracy.

[0132] 1.4 The industrial control computer processes and analyzes the images captured by the camera module in real time. The specific implementation process is as follows:

[0133] 1.4.1 The rule compilation module has set the corresponding inspection requirements for the differential housing, including: key area division (the bearing shaft diameter of the differential housing, the main reduction gear mounting stop and the differential housing window are generally key areas), defect category, defect size requirements, and defect quantity requirements;

[0134] 1.4.2 The specific detection process executed by the image processing module is as follows: Based on feature extraction and matching, image features (such as SIFT, SURF, ORB, etc.) are extracted using an image processing library. The features of standard defect-free samples and standard defective samples are compared to identify the corresponding defects. The number of pixels is calculated to confirm the current defect size and area.

[0135] 1.5 After the data processing module calculates the specific value, it compares it with the detection requirements of different areas. If the requirements are met, the corresponding data is transferred to the storage module for storage and bound to the barcode scanning information. If the requirements are not met, an error value is fed back to the feedback module, triggering an alarm and stopping the station.

[0136] 1.6 The self-learning module records each image identified as a defect into the feature database, improves the database, and enhances the accuracy of defect identification.

[0137] 1.7 After the inspection is completed, the differential assembly is removed. The gravity sensor in the inspection station recognizes that the workpiece has been removed, the inspection program is reset, and the machine waits for the next inspection.

[0138] 1.8 The detection system automatically compiles defect results for a certain period based on preset defect categories and areas, which is used to support production line problem investigation and optimization.

[0139] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0140] Based on the same inventive concept, this application also provides a differential housing defect detection device for implementing the differential housing defect detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the differential housing defect detection device provided below can be found in the limitations of the differential housing defect detection method described above, and will not be repeated here.

[0141] In one exemplary embodiment, such as Figure 5 As shown, a differential housing defect detection device is provided. The device includes a detection station, a defect detection module, and a data update module. The detection station includes an installation positioning module, an image acquisition module, a barcode scanning module, and a light source module.

[0142] The installation includes a positioning module for locating the differential housing; a light source module and a barcode scanning module for identifying and recording the serial number information of the differential housing.

[0143] The image acquisition module is used to acquire external images of the differential housing from different angles;

[0144] The defect detection module is used to perform real-time defect detection on external images from different angles based on a preset defect recognition dataset and output the defect detection results.

[0145] The data update module is used to adjust the preset defect identification dataset based on the defect detection results.

[0146] In an exemplary embodiment, the image acquisition module is specifically used to acquire external images of at least one standard defect-free sample, construct a preset defect identification dataset, which includes standard comparison images and a defect learning set; calibrate the camera module by performing data calibration using a fixed-size scale to determine the ratio of image pixels to actual size.

[0147] In one exemplary embodiment, the defect detection module is configured to: store the defect detection result and its corresponding number information if the defect detection result meets the requirements; and report an error value and issue an alarm if the defect detection result does not meet the requirements.

[0148] In one exemplary embodiment, the differential housing is divided into key areas and non-key areas, the key areas including the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window;

[0149] The image acquisition module is specifically used to acquire external images from different angles for key areas and other areas, with the image acquisition frequency for key areas being higher than that for other areas.

[0150] In one exemplary embodiment, the inspection station integrates an installation positioning module, a barcode scanning module, and a light source module; the installation positioning module includes a clamping unit for assembling the differential housing onto the inspection station at a fixed position and a fixed angle.

[0151] In an exemplary embodiment, the defect detection module is specifically used to acquire image features of external images from different angles and convert the image features from an image format into a data format; based on the image features converted into a data format and a preset defect recognition dataset, identify the defect type and defect size; and output the defect detection result according to the defect type and defect size.

[0152] Each module in the aforementioned differential housing defect detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0153] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a method for detecting defects in a differential housing. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0154] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0155] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0156] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0157] Acquire external images of the differential housing from different angles;

[0158] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0159] Adjust the preset defect identification dataset based on the defect detection results.

[0160] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0161] Collect external images of at least one standard defect-free sample and construct a pre-defined defect identification dataset, which includes standard comparison images and a defect learning set.

[0162] The camera module is calibrated by using a ruler of a fixed size to perform data calibration and determine the ratio of image pixels to their actual size.

[0163] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0164] If the defect detection results meet the requirements, the defect detection results and corresponding number information will be stored.

[0165] If the defect detection results do not meet the requirements, an error value will be reported and an alarm will be triggered.

[0166] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0167] Divide the differential housing into key areas and non-key areas. Key areas include the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window.

[0168] External images from different angles were collected for both key areas and other areas, with the image collection frequency for key areas being higher than that for other areas.

[0169] In one embodiment, the inspection station integrates a positioning module, a barcode scanning module, and a light source module.

[0170] The mounting and positioning module includes a clamping unit, which is used to assemble the differential housing into the testing station at a fixed position and angle.

[0171] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0172] Acquire image features from external images at different angles, and convert the image features from image format to data format;

[0173] Based on image features converted to data format and a pre-set defect recognition dataset, the defect type and defect size are identified.

[0174] Output the defect detection results based on the defect type and defect size.

[0175] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0176] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0177] Acquire external images of the differential housing from different angles;

[0178] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0179] Adjust the preset defect identification dataset based on the defect detection results.

[0180] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0181] Collect external images of at least one standard defect-free sample and construct a pre-defined defect identification dataset, which includes standard comparison images and a defect learning set.

[0182] The camera module is calibrated by using a ruler of a fixed size to perform data calibration and determine the ratio of image pixels to their actual size.

[0183] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0184] If the defect detection results meet the requirements, the defect detection results and corresponding number information will be stored.

[0185] If the defect detection results do not meet the requirements, an error value will be reported and an alarm will be triggered.

[0186] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0187] Divide the differential housing into key areas and non-key areas. Key areas include the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window.

[0188] External images from different angles were collected for both key areas and other areas, with the image collection frequency for key areas being higher than that for other areas.

[0189] In one embodiment, the inspection station integrates a positioning module, a barcode scanning module, and a light source module.

[0190] The mounting and positioning module includes a clamping unit, which is used to assemble the differential housing into the testing station at a fixed position and angle.

[0191] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0192] Acquire image features from external images at different angles, and convert the image features from image format to data format;

[0193] Based on image features converted to data format and a pre-set defect recognition dataset, the defect type and defect size are identified.

[0194] Output the defect detection results based on the defect type and defect size.

[0195] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0196] The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing.

[0197] Acquire external images of the differential housing from different angles;

[0198] Based on a pre-set defect recognition dataset, defects are detected in external images from different angles, and the defect detection results are output.

[0199] Adjust the preset defect identification dataset based on the defect detection results.

[0200] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0201] Collect external images of at least one standard defect-free sample and construct a pre-defined defect identification dataset, which includes standard comparison images and a defect learning set.

[0202] The camera module is calibrated by using a ruler of a fixed size to perform data calibration and determine the ratio of image pixels to their actual size.

[0203] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0204] If the defect detection results meet the requirements, the defect detection results and corresponding number information will be stored.

[0205] If the defect detection results do not meet the requirements, an error value will be reported and an alarm will be triggered.

[0206] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0207] Divide the differential housing into key areas and non-key areas. Key areas include the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window.

[0208] External images from different angles were collected for both key areas and other areas, with the image collection frequency for key areas being higher than that for other areas.

[0209] In one embodiment, the inspection station integrates a positioning module, a barcode scanning module, and a light source module.

[0210] The mounting and positioning module includes a clamping unit, which is used to assemble the differential housing into the testing station at a fixed position and angle.

[0211] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0212] Acquire image features from external images at different angles, and convert the image features from image format to data format;

[0213] Based on image features converted to data format and a pre-set defect recognition dataset, the defect type and defect size are identified.

[0214] Output the defect detection results based on the defect type and defect size.

[0215] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0216] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0217] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0218] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting defects in a differential housing, characterized in that, The method includes: At least one standard, defect-free differential housing sample is acquired to form an external image, a standard comparison pattern is constructed, and defect sample images with pre-labeled defect types are acquired to form a defect learning set. The standard comparison pattern and the defect learning set are then combined to form a preset defect identification dataset. The camera module is calibrated by using a ruler of a fixed size to perform data calibration and determine the ratio of image pixels to their actual size. The differential housing is positioned using the installation and positioning module at the inspection station. The light source module and the barcode scanning module are then activated to identify and record the serial number information of the differential housing. A camera module, which is arranged around the differential housing and configured to acquire external images of the differential housing from different angles without rotating the differential housing, is activated. According to the key areas and non-key areas of the differential housing divided by preset rules, external images of the differential housing from different angles are acquired. The key areas include the differential housing bearing shaft diameter, the main reduction gear mounting stop, and the differential housing window. Based on a preset defect identification dataset, defect detection is performed on external images from different angles, and defect detection results are output. The defect detection includes: extracting image features from the external image, comparing the image features with features in the preset defect identification dataset, identifying the defect type, and calculating the defect size based on the ratio of the image pixels to the actual size. Based on the defect detection results, the images identified as defects and their features are entered into the defect learning set to update the preset defect recognition dataset.

2. The method according to claim 1, characterized in that, Following the output of the defect detection results, the following is also included: If the defect detection results meet the requirements, the defect detection results and the corresponding number information are stored. If the defect detection results do not meet the requirements, an error value will be reported and an alarm will be triggered.

3. The method according to claim 1, characterized in that, Acquire external images of the differential housing from different angles, including: Divide the differential housing into key and non-key areas; External images from different angles were collected for both key areas and other areas, with the image collection frequency for key areas being higher than that for other areas.

4. The method according to claim 1, characterized in that, The testing station integrates a positioning module, a barcode scanning module, and a light source module. The installation and positioning module includes a clamping unit, which is used to assemble the differential housing onto the testing station at a fixed position and a fixed angle.

5. The method according to claim 1, characterized in that, The method of detecting defects in external images from different angles based on a preset defect recognition dataset and outputting defect detection results includes: Acquire image features from external images at different angles, and convert the image features from image format to data format; Based on image features converted to data format and a pre-set defect recognition dataset, the defect type and defect size are identified. Output the defect detection results based on the defect type and defect size.

6. A differential housing defect detection device, characterized in that, The apparatus, using the method of any one of claims 1-5, comprises a detection station, a defect detection module, and a data update module. The detection station includes an installation and positioning module, an image acquisition module, a barcode scanning module, and a light source module. The installation positioning module is used to position the differential housing; the light source module and the barcode scanning module are used to identify and record the serial number information of the differential housing. The image acquisition module is used to acquire external images of the differential housing from different angles; The defect detection module is used to perform defect detection on external images from different angles in real time based on a preset defect recognition dataset and output the defect detection results. The data update module is used to adjust the preset defect identification dataset based on the defect detection results.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images