Detection data processing method and device, computer device and readable storage medium

Abstract

The application relates to a detection data processing method and device, computer equipment and a readable storage medium. The method comprises the following steps: determining a to-be-detected problem item corresponding to a device category of a to-be-detected device according to the device category, wherein the to-be-detected problem item comprises an appearance problem item and a function problem item; acquiring appearance image data and device detection data of the to-be-detected device; performing image recognition on the appearance image data to obtain an appearance recognition result of the appearance problem item; performing data analysis on the device detection data to obtain a function recognition result of the function problem item; and mapping the appearance recognition result and the function recognition result to a predetermined quality inspection standard template to generate a quality inspection report of the to-be-detected device. The method can improve the quality inspection efficiency of second-hand 3C products.

Description

Technical Field

[0001] This application relates to the field of product quality inspection technology, and in particular to a method, apparatus, computer equipment, and readable storage medium for processing test data. Background Technology

[0002] With the development of the 3C product (Computer, Communication, Consumer Electronics, also known as "information appliances") industry, terminal products such as computers, mobile phones, tablets, digital cameras, and smart home devices are being updated and iterated at an increasingly rapid pace, and the market for second-hand 3C products has emerged accordingly.

[0003] In the secondhand 3C product industry, quality inspection is a crucial step in ensuring product quality and maintaining trust in transactions. Currently, the traditional quality inspection process largely relies on inspectors manually entering the results after completing each inspection item, resulting in low efficiency for secondhand 3C product quality inspection. Summary of the Invention

[0004] Therefore, it is necessary to provide a data processing method, apparatus, computer equipment, and readable storage medium that can improve the quality inspection efficiency of second-hand 3C products in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a method for processing detection data, including:

[0006] Based on the equipment category of the equipment to be inspected, determine the corresponding inspection issues for that equipment category, including appearance issues and functional issues.

[0007] Acquire external image data and equipment inspection data of the equipment to be inspected;

[0008] Image recognition is performed on the appearance image data to obtain the appearance recognition results of appearance problem items;

[0009] Data analysis is performed on the equipment test data to obtain the functional identification results of the functional problem items;

[0010] The appearance recognition results and function recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0011] In one embodiment, image recognition is performed on the appearance image data to obtain the appearance recognition result of the appearance problem item, including:

[0012] Determine the image recognition model for appearance issues;

[0013] The appearance image data is processed by an image recognition model to obtain the appearance recognition results of appearance problem items.

[0014] In one embodiment, the image recognition model includes a feature extraction network. The image recognition model processes appearance image data to obtain appearance recognition results for appearance problem items, including:

[0015] Feature extraction networks are used to extract features from appearance image data to obtain appearance image features;

[0016] The appearance image features are compared with the appearance defect features in a predetermined defect feature library to obtain the appearance defect features that match the appearance image features;

[0017] Based on appearance defect features matched with appearance image features, appearance recognition results for appearance problem items are generated.

[0018] In one embodiment, the method further includes:

[0019] Acquire misjudged image data and label information for the misjudged image data, wherein the misjudged image data includes appearance image data in which appearance recognition results are misjudged;

[0020] The misjudged image data and its label information are used as supplementary sample data and added to the original sample training set of the image recognition model to obtain a new sample training set.

[0021] The image recognition model is retrained based on the new sample training set to obtain a new image recognition model.

[0022] In one embodiment, the device detection data is parsed to obtain the functional identification results of the functional problem items, including:

[0023] Data parsing is performed on the equipment testing data to obtain the testing result fields;

[0024] The detection result fields are mapped to the corresponding functional problem items to obtain the functional identification results of the functional problem items.

[0025] In one embodiment, before mapping the appearance recognition results and function recognition results to a predetermined quality inspection standard template to generate a quality inspection report for the device to be inspected, the method further includes:

[0026] Obtain historical identification results information for device categories, including historical appearance results and historical function results;

[0027] If the difference between the appearance recognition result and the historical appearance result is greater than or equal to a predetermined appearance difference threshold, or if the difference between the function recognition result and the historical function result is greater than or equal to a predetermined function difference threshold, an abnormal prompt message will be output.

[0028] In response to the confirmation operation for the abnormal prompt information, the steps of mapping the appearance recognition results and function recognition results to the predetermined quality inspection standard template are performed to generate a quality inspection report for the equipment to be inspected.

[0029] In response to the adjustment operation for the abnormal prompt information, after adjusting the appearance recognition results and function recognition results according to the adjustment operation, the step of mapping the appearance recognition results and function recognition results to the predetermined quality inspection standard template is executed to generate the quality inspection report of the equipment to be inspected.

[0030] In one embodiment, the method further includes:

[0031] In response to the associated operation for a specified equipment category, determine the problem items to be detected corresponding to the specified equipment category;

[0032] Based on the issues to be tested corresponding to the specified equipment category, construct a mapping relationship between the specified equipment category and the issues to be tested.

[0033] Secondly, this application also provides a detection data processing device, comprising:

[0034] The problem item determination module is used to determine the problem items to be tested corresponding to the equipment category of the equipment to be tested, including appearance problem items and functional problem items.

[0035] The data acquisition module is used to acquire the appearance image data and equipment inspection data of the equipment under inspection;

[0036] The appearance recognition module is used to perform image recognition on appearance image data to obtain the appearance recognition results of appearance problem items;

[0037] The function identification module is used to parse the equipment test data and obtain the function identification results of the functional problem items.

[0038] The report generation module is used to map the appearance recognition results and function recognition results to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0039] 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:

[0040] Based on the equipment category of the equipment to be inspected, determine the corresponding inspection issues for that equipment category, including appearance issues and functional issues.

[0041] Acquire external image data and equipment inspection data of the equipment to be inspected;

[0042] Image recognition is performed on the appearance image data to obtain the appearance recognition results of appearance problem items;

[0043] Data analysis is performed on the equipment test data to obtain the functional identification results of the functional problem items;

[0044] The appearance recognition results and function recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0045] 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:

[0046] Based on the equipment category of the equipment to be inspected, determine the corresponding inspection issues for that equipment category, including appearance issues and functional issues.

[0047] Acquire external image data and equipment inspection data of the equipment to be inspected;

[0048] Image recognition is performed on the appearance image data to obtain the appearance recognition results of appearance problem items;

[0049] Data analysis is performed on the equipment test data to obtain the functional identification results of the functional problem items;

[0050] The appearance recognition results and function recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0051] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0052] Based on the equipment category of the equipment to be inspected, determine the corresponding inspection issues for that equipment category, including appearance issues and functional issues.

[0053] Acquire external image data and equipment inspection data of the equipment to be inspected;

[0054] Image recognition is performed on the appearance image data to obtain the appearance recognition results of appearance problem items;

[0055] Data analysis is performed on the equipment test data to obtain the functional identification results of the functional problem items;

[0056] The appearance recognition results and function recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0057] The aforementioned data processing method, apparatus, computer equipment, and readable storage medium determine the corresponding inspection issues based on the equipment category of the equipment to be inspected. These issues include appearance and functional problems. The method acquires appearance image data and equipment inspection data, performs image recognition on the appearance image data to obtain appearance recognition results for the appearance issues, and parses the equipment inspection data to obtain functional recognition results for the functional problems. Finally, it maps the appearance and functional recognition results to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected. This application addresses the different appearance and functional issues corresponding to different equipment categories by conducting inspections specifically for those categories, ensuring clear and non-redundant inspection objectives and adapting to the differentiated quality inspection needs of various equipment categories. Furthermore, based on the appearance image data and equipment inspection data of the equipment under inspection, the appearance recognition results of appearance problems and the function recognition results of function problems are identified and mapped to the predetermined quality inspection standard template. This can reduce the misjudgment and omission of manual visual inspection and the influence of subjective experience. Compared with the method of manually analyzing and filling in each item one by one, it helps to improve the efficiency of quality inspection. Attached Figure Description

[0058] 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.

[0059] Figure 1 This is a diagram illustrating the application environment of a detection data processing method in one embodiment;

[0060] Figure 2 This is a flowchart illustrating a detection data processing method in one embodiment;

[0061] Figure 3 This is a schematic diagram of the display interface for a problem item to be detected, as shown in one embodiment.

[0062] Figure 4 This is a schematic diagram of the display interface for a functional issue item in one embodiment;

[0063] Figure 5 This is a schematic diagram of a portion of the display interface of a quality inspection report, as shown in one embodiment.

[0064] Figure 6This is a flowchart illustrating the detection data processing method in another embodiment;

[0065] Figure 7 This is a schematic diagram of the display interface for a problem item to be detected, as shown in one embodiment.

[0066] Figure 8 This is a flowchart illustrating the detection data processing method in yet another embodiment;

[0067] Figure 9 This is a structural block diagram of a detection data processing device in one embodiment;

[0068] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0069] 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.

[0070] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0071] The detection data processing method provided in this application embodiment can be applied to, for example... Figure 1In 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 placed on a cloud or other network server. Server 104 can send device information of the device to be inspected to terminal 102. After receiving the device information, terminal 102 determines the corresponding inspection issues based on the device category, including appearance and function issues. It acquires appearance image data and device inspection data of the device to be inspected, performs image recognition on the appearance image data to obtain the appearance recognition result for the appearance issues, and performs data parsing on the device inspection data to obtain the function recognition result for the function issues. The appearance and function recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the device to be inspected. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and other terminal devices. 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.

[0072] In one exemplary embodiment, such as Figure 2 As shown, a detection data processing method is provided, which is applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps S110 to S150. Wherein:

[0073] Step S110: Based on the equipment category of the equipment to be inspected, determine the corresponding inspection issues for the equipment category, including appearance issues and functional issues.

[0074] The devices to be inspected can be second-hand 3C products that are expected to undergo quality inspection. The device category refers to the classification of second-hand 3C products, such as laptops, mobile phones, and tablets. The inspection items include appearance issues and functional issues. Appearance issues are options used to describe the appearance of the device category, such as the surface condition of the body and the degree of bending of the body. Functional issues are options used to describe the functions of the device category, such as the display function status, touch function status, gyroscope function status, and button function status.

[0075] This embodiment can pre-set corresponding inspection issues for different device categories of second-hand 3C products (such as mobile phones, tablets, laptops, etc.), including appearance issues (such as scratches, abnormal screen display, malfunctions, etc.) and functional issues (such as button functions, charging functions, camera functions, network connectivity functions, etc.), establishing an association mapping relationship between device categories and inspection issues. For example, device categories and their corresponding inspection issues can be added, deleted, or modified and stored in the system database to provide a standardized basis for subsequent testing. Furthermore, considering that some inspection items may require subjective evaluation, this embodiment can also set priorities for the inspection items. For example, inspection items with a priority higher than or equal to a specified level will be automatically identified via step S120, while inspection items with a priority lower than the specified level will be manually confirmed.

[0076] Upon receiving a quality inspection task, this embodiment determines the equipment information of the equipment to be inspected, such as equipment category, serial number, brand, and model. Then, based on this equipment information, it determines the equipment category and queries the mapping relationship between the equipment category and the issues to be inspected, obtaining the corresponding issues for that equipment category. These issues include appearance issues and functional issues.

[0077] In some embodiments, the method further includes:

[0078] Step S210: In response to the association operation for the specified equipment category, determine the problem items to be detected corresponding to the specified equipment category;

[0079] Step S220: Based on the problem items to be detected corresponding to the specified equipment category, construct a mapping relationship between the specified equipment category and the problem items to be detected.

[0080] Among them, the association operation refers to operations such as adding / deleting / modifying that indicate the association between a specified equipment category and the problem item to be detected.

[0081] like Figure 3 As shown, Figure 3 This is a schematic diagram illustrating the interface for constructing the mapping relationship between a specified equipment category and the issues to be tested. This embodiment can respond to association operations such as adding, deleting, and modifying candidate issues for a specified equipment category to determine the issues to be tested corresponding to that specified equipment category. Furthermore, based on the issues to be tested corresponding to the specified equipment category, a mapping relationship between the specified equipment category and the issues to be tested can be constructed, thereby providing a standardized basis for the required issues to be tested for the equipment category of the equipment to be tested during subsequent testing.

[0082] Step S120: Obtain the appearance image data and equipment inspection data of the device under inspection;

[0083] The external image data consists of images taken of the equipment under inspection, while the equipment inspection data consists of feedback data obtained from testing the equipment's functions. The external image data can include images of the front, back, sides, corners, interfaces, marking areas, assembly gaps, and areas prone to defects—all parts of the equipment to be inspected. The equipment inspection data can include test data for the equipment's display, touch, button, and speaker functions.

[0084] For example, when the terminal, which is the subject of this embodiment, has both a camera component and a detection component, the camera component of the terminal can be used to capture an image of the appearance of the device under test, obtaining appearance image data. The detection component of the terminal can then be used to detect the function of the device under test, obtaining device detection data. When the terminal, which is the subject of this embodiment, has a camera component but no detection component, the camera component of the terminal can be used to capture an image of the appearance of the device under test, obtaining appearance image data. This image data can then be obtained by establishing a communication connection with the detection device that detects the function of the device under test. Alternatively, the appearance image data of the device under test can be obtained by establishing a communication connection with the camera component that captures an image of the appearance of the device under test.

[0085] Step S130: Perform image recognition on the appearance image data to obtain the appearance recognition result of the appearance problem item;

[0086] In this embodiment, the appearance image data can be preprocessed to obtain a preprocessed appearance image. Preprocessing may include image denoising, image enhancement, image cropping, normalization, and image grayscale conversion (binarization). Furthermore, this embodiment can use an image recognition model to perform image recognition on the appearance image data to obtain appearance identification results for appearance issues. The image recognition model is a deep learning model capable of identifying appearance defects corresponding to appearance issues, such as ResNet (Residual Network), MobileNet (Mobile Network), and YOLO (YouOnly Look Once). In another embodiment, this embodiment can perform feature extraction on the preprocessed appearance image to obtain appearance image features. These appearance image features correspond to appearance issues; that is, the extracted features are the appearance image features corresponding to the appearance issues of this equipment category. For example, for appearance issues (scratches, dents, defects), contour features, edge features, and texture abrupt change features can be extracted from the appearance image data. For instance, scratches are represented by linear texture abrupt changes in the image, and dents are represented by irregular defects in the corner contours, for subsequent accurate identification. For appearance issues (color difference, dissimilar colors), color features and pixel grayscale value features can be extracted from the appearance image data, thereby identifying dissimilar pixels and color blocks outside the preset standard color value range by comparing them. For appearance issues (body bending), geometric features can be extracted from the appearance image data, thereby determining the body condition through the edge contours and geometric shapes of the device under inspection. Then, the appearance image features are compared with appearance defect features in a predetermined defect feature library to obtain appearance defect features that match the appearance image features. Based on the appearance defect features that match the appearance image features, the appearance recognition result for the appearance issue is generated. This embodiment decouples feature extraction from feature comparison, thereby allowing for adaptive adjustment of appearance defect features in a predetermined defect feature library according to actual needs. Compared to an end-to-end holistic model, the decoupling scheme is more adaptable to scenarios with fewer samples and has stronger interpretability.

[0087] Step S140: Analyze the equipment test data to obtain the functional identification results of the functional problem items;

[0088] This embodiment can parse the equipment test data to obtain a test result field. Then, based on the mapping relationship between the test result field and the functional problem item, the test result field is mapped to the corresponding functional problem item to obtain the functional identification result of the functional problem item. In some examples, the test result field is a test value, but the functional identification result is a conclusive result. In such cases, the test value can be compared with the corresponding discrimination threshold to obtain the test result. Then, the test result corresponding to the test result field is mapped to the corresponding functional problem item to obtain the functional identification result of the functional problem item.

[0089] In some embodiments, data parsing of device detection data is performed to obtain functional identification results for functional problem items, including:

[0090] Step S141: Parse the equipment test data to obtain the test result field;

[0091] Step S142: Map the detection result field to the corresponding functional problem item to obtain the functional identification result of the functional problem item.

[0092] For example, this embodiment can parse the device detection data to obtain a detection result field, and then map the detection result field to the corresponding functional problem item according to the mapping relationship between the detection result field and the functional problem item, thereby obtaining the functional identification result of the functional problem item. Figure 4 As shown, in some examples where the detection result field is a detection value, but the function identification result is a conclusive result, the detection value can be compared with the corresponding discrimination threshold to obtain the detection result. Then, the detection result corresponding to the detection result field is mapped to the corresponding functional problem item to obtain the function identification result of the functional problem item. For example, if data parsing of device detection data yields a detection result field of "Battery Health 40%", then 40% can be compared with the discrimination threshold of 50% corresponding to "Poor Health", resulting in a detection result of "Unqualified". The detection result "Unqualified" corresponding to the detection result field is then mapped to the corresponding functional problem item "Battery Health", resulting in the function identification result "Unqualified" for the functional problem item "Battery Health".

[0093] Step S150: Map the appearance recognition results and function recognition results to the predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0094] Among them, the pre-set quality inspection standard template is a pre-set standard template for quality inspection reports.

[0095] This embodiment can pre-establish a mapping relationship between each template field in a predetermined quality inspection standard template and the problem items to be inspected. Therefore, after obtaining the appearance identification results for appearance problems and the functional identification results for functional problems, this embodiment can map the appearance identification results and functional identification results to the corresponding template fields in the predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected. For example... Figure 5 As shown, this embodiment can also fill the equipment information of the device to be inspected into a predetermined quality inspection standard template to improve the description of the basic information of the device to be inspected. For example, the field structure of the template is predefined (such as the template containing fields such as "basic product information (category, serial number) - appearance inspection items - functional inspection items - comprehensive judgment result - quality inspector signature") to obtain the predetermined quality inspection standard template, and then establish a mapping relationship between the fields of the problem items to be inspected and the template fields of the predetermined quality inspection standard template, such as the field of the problem item "screen scratch level" corresponding to the "appearance inspection - screen defect - level" field in the template). If there are unmatched problem items to be inspected (such as exclusive inspection items for customized equipment categories), the problem item to be inspected is marked as "to be supplemented" to remind the staff to improve the problem item to be inspected, and finally generate a complete quality inspection report.

[0096] In addition, this embodiment can also generate visual reports (such as line charts and bar charts) and display the visual reports to show the quality inspection efficiency indicators (such as the average quality inspection time of a single product, the success rate of reading the recognition result, etc.) and error rate indicators (such as the manual review and correction rate, the matching error rate of the report field, etc.) based on the quality inspection records of the equipment to be inspected within a specified time period.

[0097] In the aforementioned data processing method, the corresponding inspection issues are determined based on the equipment category of the equipment to be inspected. These issues include appearance issues and functional issues. The method acquires appearance image data and equipment inspection data of the equipment to be inspected. Image recognition is performed on the appearance image data to obtain the appearance recognition results for the appearance issues. Data parsing is performed on the equipment inspection data to obtain the functional recognition results for the functional issues. The appearance recognition results and functional recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected. This embodiment of the application addresses the different appearance and functional issues corresponding to different equipment categories, ensuring clear and non-redundant inspection targets and adapting to the differentiated quality inspection needs of different equipment categories. Furthermore, by identifying the appearance recognition results for appearance issues and the functional recognition results for functional issues based on the appearance image data and equipment inspection data of the equipment to be inspected, and mapping them to a predetermined quality inspection standard template, the method reduces misjudgments and omissions caused by manual visual inspection and the influence of subjective experience. Compared to manual parsing and item-by-item filling, this method helps improve quality inspection efficiency.

[0098] In one exemplary embodiment, such as Figure 6 As shown, image recognition is performed on the appearance image data to obtain the appearance recognition result of the appearance problem item, including steps S310 to S320. Wherein:

[0099] Step S310: Determine the image recognition model for appearance problem items;

[0100] Step S320: Process the appearance image data using an image recognition model to obtain the appearance recognition result for the appearance problem item.

[0101] Among them, the image recognition model is a deep learning model that can identify appearance defects corresponding to appearance problem items, such as pre-trained ResNet (Residual Network), MobileNet (Mobile Network), YOLO (You Only Look Once) detector, etc.

[0102] Because the evaluation standards for the appearance of different used 3C products vary, this embodiment can select different image recognition models for different device categories. Of course, this embodiment can also select corresponding image recognition models for different appearance issues. This embodiment can select the appropriate image recognition model based on the device category and the appearance issue. Then, the appearance image data can be processed using the image recognition model to obtain the appearance recognition results for the appearance issue. For example... Figure 7 As shown, after obtaining the appearance recognition results of the appearance problem item, the appearance recognition result of the appearance problem item "degree of body bending" "no body bending" can be displayed on the display interface.

[0103] This embodiment determines an image recognition model for appearance problems, processes appearance image data using the image recognition model, and obtains appearance recognition results for appearance problems, which can further improve the accuracy of appearance recognition results.

[0104] In some embodiments, the image recognition model includes a feature extraction network. The image recognition model processes appearance image data to obtain appearance recognition results for appearance problem items, including:

[0105] Step S321: Extract features from the appearance image data using a feature extraction network to obtain appearance image features;

[0106] Step S322: Compare the appearance image features with the appearance defect features in the predetermined defect feature library to obtain the appearance defect features that match the appearance image features.

[0107] Step S323: Based on the appearance defect features matched with the appearance image features, generate the appearance recognition result for the appearance problem item.

[0108] The image recognition model includes a feature extraction network, which is a deep learning network used to extract features from image data, such as a convolutional neural network or a Transformer. The predefined defect feature library includes appearance defect features corresponding to several appearance defects.

[0109] This embodiment uses a feature extraction network to extract features from appearance image data, obtaining appearance image features that correspond to appearance problem items. Specifically, the extracted features are those corresponding to the appearance problem items for that equipment category. For example, for appearance problem items (scratches, dents, defects), contour features, edge features, and texture abrupt change features can be extracted from the appearance image data. For instance, scratches are represented by linear texture abrupt changes in the image, and dents are represented by irregular defects in the corner contours, for accurate subsequent identification. For appearance problem items (color difference, dissimilar colors), color features and pixel grayscale value features can be extracted from the appearance image data, allowing for the identification of dissimilar pixels and color blocks outside a preset standard color value range. For appearance problem items (body bending), geometric features can be extracted from the appearance image data, allowing the determination of the body condition through the edge contours and geometric shapes of the equipment under inspection. The appearance image features are then compared with appearance defect features in a predetermined defect feature library to obtain appearance defect features that match the appearance image features. Therefore, this embodiment can determine the appearance defect corresponding to the appearance defect feature matching the appearance image feature as the appearance recognition result of the appearance problem item. This embodiment decouples feature extraction from feature comparison, thereby allowing for adaptive adjustment of appearance defect features in a predetermined defect feature library according to actual needs. Compared to an end-to-end holistic model, the decoupling scheme is more adaptable to scenarios with fewer samples and has stronger interpretability.

[0110] In some embodiments, the method further includes:

[0111] Step S410: Obtain misjudged image data and label information of misjudged image data, wherein misjudged image data includes appearance image data in which appearance recognition results are misjudged;

[0112] Step S420: The misjudged image data and the label information of the misjudged image data are added as supplementary sample data to the original sample training set of the image recognition model to obtain a new sample training set.

[0113] Step S430: Retrain the image recognition model based on the new sample training set to obtain a new image recognition model.

[0114] Among them, the misjudged image data includes appearance image data in which the appearance recognition results are misjudged, and the label information is the true appearance result of the misjudged image data.

[0115] To ensure the accuracy of the image recognition model, this embodiment can acquire misjudged image data and its label information. The misjudged image data includes appearance images where misjudgments have occurred. The misjudged image data and its label information are added as supplementary sample data to the original sample training set of the image recognition model, resulting in a new sample training set. This embodiment can then train the image recognition model in its initial state based on the new sample training set, obtaining a new image recognition model. Thus, by updating the image recognition model, the recognition accuracy of the image recognition model is ensured.

[0116] In one exemplary embodiment, such as Figure 8 As shown, before mapping the appearance recognition results and function recognition results to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected, the method further includes steps S510 to S540. Wherein:

[0117] Step S510: Obtain historical identification result information for the device category, wherein the historical identification result information includes historical appearance results and historical function results;

[0118] Step S520: If the difference between the appearance recognition result and the historical appearance result is greater than or equal to a predetermined appearance difference threshold, or the difference between the function recognition result and the historical function result is greater than or equal to a predetermined function difference threshold, output an abnormal prompt message.

[0119] Step S530: In response to the confirmation operation for the abnormal prompt information, perform the step of mapping the appearance recognition result and function recognition result to the predetermined quality inspection standard template and generating a quality inspection report for the equipment to be inspected.

[0120] Step S540: In response to the adjustment operation for the abnormal prompt information, after adjusting the appearance recognition result and function recognition result according to the adjustment operation, the step of mapping the appearance recognition result and function recognition result to the predetermined quality inspection standard template and generating the quality inspection report of the device to be inspected is executed.

[0121] The historical identification results information can include historical appearance results and historical function results for different time periods within the equipment category. For example, historical appearance results could be the appearance identification result with the highest probability during a historical period, and historical function results could be the function identification result with the highest probability during a historical period. The difference between the appearance identification results and historical appearance results can be a conclusive statement describing whether the appearance identification results are consistent with historical appearance results, or a statement describing the difference or ratio between the result representation values ​​of the appearance identification results and historical appearance results. Similarly, the difference between the function identification results and historical function results can be a conclusive statement describing whether the function identification results are consistent with historical function results, or a statement describing the difference or ratio between the result representation values ​​of the function identification results and historical function results.

[0122] This embodiment can acquire historical identification result information for device categories, including historical appearance results and historical function results. When the difference between the appearance identification result and the historical appearance result is greater than or equal to a predetermined appearance difference threshold, or the difference between the function identification result and the historical function result is greater than or equal to a predetermined function difference threshold, an anomaly prompt message is output. For example, if the camera detection pass rate for a certain model of mobile phone was 90% during a historical period, the historical function result could be "normal." If the function identification result is "abnormal," and the result is inconsistent with the historical function result, it can be determined that the difference between the function identification result and the historical function result is greater than or equal to a predetermined function difference threshold, and an anomaly prompt message is output to remind staff to conduct a second review, reducing the risk of misjudgment. If the staff's second review is correct, they can perform a confirmation operation (such as clicking the "OK" button) based on the anomaly prompt message. In response to the confirmation operation on the anomaly prompt message, this embodiment can execute the step of mapping the appearance identification result and the function identification result to a predetermined quality inspection standard template to generate a quality inspection report for the device to be inspected. If the staff's second review contains a misjudgment, they can adjust the appearance identification result and the function identification result based on the anomaly prompt message (such as modifying the appearance identification result or the function identification result). This embodiment can respond to the adjustment operation for the abnormal prompt information. After adjusting the appearance recognition result and function recognition result according to the adjustment operation, it can perform the step of mapping the appearance recognition result and function recognition result to the predetermined quality inspection standard template and generating the quality inspection report of the device to be inspected.

[0123] Therefore, this embodiment can compare the current appearance recognition results and function recognition results with the historical appearance results and function results before officially generating the quality inspection report of the device to be inspected, thereby reviewing the results when there are significant differences, which helps to reduce the probability of misjudgment.

[0124] As a specific example, in this embodiment, the server can create a quality inspection task and then assign it to a terminal. Upon receiving the quality inspection task, the terminal reads the image recognition results for appearance issues and the functional test results for functional issues, corresponding to the product to be inspected. For any issue (appearance or functional) without a test result, supplementary prompts can be output. In response to the supplementary prompts, supplementary test information is obtained and used as the test result for that issue. Complete test data (appearance recognition results and functional recognition results) can be synchronized to the server via an encryption protocol. The server's database stores data using an associated index to ensure traceability. The server can map the complete test data to a predetermined quality inspection standard template to generate a quality inspection report for the device to be inspected. If there are unmatched template fields in the predetermined quality inspection standard template, a field missing prompt can be output. In response to the field missing prompt, supplementary field information is obtained and mapped to the unmatched template fields to obtain the final quality inspection report. This embodiment can also periodically collect quality inspection indicators such as inspection efficiency and error rate, add misjudged detection samples to the sample training set of the image recognition model to optimize model performance, and adjust the configuration of the problem items to be detected and the mapping rules of the predetermined quality inspection standard template according to business needs.

[0125] It should be understood that although the steps in the flowcharts of the above embodiments 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 above embodiments 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 in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0126] Based on the same inventive concept, this application also provides a detection data processing apparatus for implementing the detection data processing method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more detection data processing apparatus embodiments provided below can be found in the limitations of the detection data processing method described above, and will not be repeated here.

[0127] In one exemplary embodiment, such as Figure 9 As shown, a detection data processing device 600 is provided, including: a problem item determination module 610, a data acquisition module 620, an appearance recognition module 630, a function recognition module 640, and a report generation module 650, wherein:

[0128] Problem item determination module 610 is used to determine the problem items to be tested corresponding to the equipment category of the equipment to be tested, wherein the problem items to be tested include appearance problem items and functional problem items;

[0129] The data acquisition module 620 is used to acquire the appearance image data and equipment inspection data of the equipment under inspection;

[0130] The appearance recognition module 630 is used to perform image recognition on appearance image data to obtain the appearance recognition result of appearance problem items;

[0131] The function identification module 640 is used to parse the equipment test data and obtain the function identification results of the functional problem items.

[0132] The report generation module 650 is used to map the appearance recognition results and function recognition results to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0133] In some embodiments, the appearance recognition module 630 is further configured to:

[0134] A model for identifying appearance problems is determined, and the appearance image data is processed using the model to obtain the appearance recognition results for the appearance problems.

[0135] In some embodiments, the image recognition model includes a feature extraction network, and the appearance recognition module 630 is further used for:

[0136] Feature extraction is performed on the appearance image data by a feature extraction network to obtain appearance image features. The appearance image features are then compared with appearance defect features in a predetermined defect feature library to obtain appearance defect features that match the appearance image features. Based on the appearance defect features that match the appearance image features, appearance recognition results for appearance problem items are generated.

[0137] In some embodiments, the detection data processing device 600 further includes a model update module, used for:

[0138] Obtain misjudged image data and its label information. The misjudged image data includes appearance images where the appearance recognition results are misjudged. Add the misjudged image data and its label information as supplementary sample data to the original sample training set of the image recognition model to obtain a new sample training set. Retrain the image recognition model based on the new sample training set to obtain a new image recognition model.

[0139] In some embodiments, the function recognition module 640 is further configured to:

[0140] The equipment test data is parsed to obtain the test result field. The test result field is then mapped to the corresponding functional problem item to obtain the functional identification result of the functional problem item.

[0141] In some embodiments, the detection data processing device 600 further includes a result verification module, used for:

[0142] The system acquires historical identification results for the equipment category, including historical appearance results and historical function results. If the difference between the appearance identification result and the historical appearance result is greater than or equal to a predetermined appearance difference threshold, or the difference between the function identification result and the historical function result is greater than or equal to a predetermined function difference threshold, an anomaly warning is output. In response to a confirmation operation for the anomaly warning, the system performs a step of mapping the appearance identification results and function identification results to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected. In response to an adjustment operation for the anomaly warning, the system adjusts the appearance identification results and function identification results according to the adjustment operation, and then performs another step of mapping the appearance identification results and function identification results to the predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

[0143] In some embodiments, the detection data processing device 600 further includes a mapping configuration module, used for:

[0144] In response to an association operation for a specified equipment category, determine the problem items to be detected corresponding to the specified equipment category, and construct a mapping relationship between the specified equipment category and the problem items to be detected based on the problem items to be detected corresponding to the specified equipment category.

[0145] Each module in the aforementioned detection data processing 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 operations corresponding to each module.

[0146] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a detection data processing method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0147] Those skilled in the art will understand that Figure 10 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.

[0148] 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 implement the steps of the embodiments described above.

[0149] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above embodiments.

[0150] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above embodiments.

[0151] 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.

[0152] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments 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.

[0153] 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.

[0154] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this 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 processing detection data, characterized in that, The method includes: Based on the equipment category of the equipment to be inspected, determine the corresponding inspection issues for that equipment category, wherein the inspection issues include appearance issues and functional issues; Acquire the appearance image data and equipment inspection data of the device under inspection; Image recognition is performed on the appearance image data to obtain the appearance recognition result of the appearance problem item; The device detection data is parsed to obtain the functional identification result of the functional problem item; The appearance recognition results and the function recognition results are mapped to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

2. The method according to claim 1, characterized in that, The step of performing image recognition on the appearance image data to obtain the appearance recognition result of the appearance problem item includes: Determine the image recognition model for the aforementioned appearance problem item; The appearance image data is processed by the image recognition model to obtain the appearance recognition result of the appearance problem item.

3. The method according to claim 2, characterized in that, The image recognition model includes a feature extraction network. The process of processing the appearance image data using the image recognition model to obtain the appearance recognition result for the appearance problem item includes: The feature extraction network is used to extract features from the appearance image data to obtain appearance image features. The appearance image features are compared with appearance defect features in a predetermined defect feature library to obtain appearance defect features that match the appearance image features; Based on the appearance defect features that match the appearance image features, the appearance recognition result of the appearance problem item is generated.

4. The method according to claim 2, characterized in that, The method further includes: Acquire misjudged image data and label information of the misjudged image data, wherein the misjudged image data includes appearance image data in which appearance recognition results are misjudged; The misjudged image data and its label information are used as supplementary sample data and added to the original sample training set of the image recognition model to obtain a new sample training set. The image recognition model is retrained based on the new sample training set to obtain a new image recognition model.

5. The method according to claim 1, characterized in that, The step of parsing the device detection data to obtain the functional identification result of the functional problem item includes: The device's detection data is parsed to obtain the detection result field; The detection result field is mapped to the corresponding functional problem item to obtain the functional identification result of the functional problem item.

6. The method according to claim 1, characterized in that, Before mapping the appearance recognition result and the function recognition result to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected, the method further includes: Obtain historical identification results information for the device category, wherein the historical identification results information includes historical appearance results and historical function results; If the difference between the appearance recognition result and the historical appearance result is greater than or equal to a predetermined appearance difference threshold, or if the difference between the function recognition result and the historical function result is greater than or equal to a predetermined function difference threshold, an abnormal prompt message will be output. In response to the confirmation operation for the abnormal prompt information, the step of mapping the appearance recognition result and the function recognition result to a predetermined quality inspection standard template and generating a quality inspection report for the device to be inspected is performed. In response to the adjustment operation for the abnormal prompt information, after adjusting the appearance recognition result and the function recognition result according to the adjustment operation, the step of mapping the appearance recognition result and the function recognition result to a predetermined quality inspection standard template and generating a quality inspection report for the device to be inspected is executed.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: In response to an association operation for a specified equipment category, determine the problem item to be detected corresponding to the specified equipment category; Based on the problem items to be detected corresponding to the specified equipment category, a mapping relationship is constructed between the specified equipment category and the problem items to be detected.

8. A detection data processing device, characterized in that, The device includes: The problem item determination module is used to determine the problem items to be tested corresponding to the equipment category of the equipment to be tested, wherein the problem items to be tested include appearance problem items and functional problem items; The data acquisition module is used to acquire the appearance image data and equipment inspection data of the device under inspection; An appearance recognition module is used to perform image recognition on the appearance image data to obtain the appearance recognition result of the appearance problem item; The function identification module is used to parse the device detection data to obtain the function identification result of the function problem item; The report generation module is used to map the appearance recognition results and the function recognition results to a predetermined quality inspection standard template to generate a quality inspection report for the equipment to be inspected.

9. 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 7.

10. 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 7.

Images