A data processing method, apparatus and device

Abstract

The application relates to a data processing method, device and equipment. The method comprises the following steps: acquiring original image data corresponding to a terminal real object; determining structured inventory data based on the original image data and a preset structured constraint; in response to a field change event being triggered, determining abnormal inventory data based on the structured inventory data and a preset inventory reference data; determining an abnormal reason and evidence materials corresponding to the abnormal inventory data based on the abnormal inventory data; performing compliance verification processing on the abnormal inventory data based on the abnormal reason and the evidence materials, and obtaining a compliance verification processing result; and if the compliance verification processing result indicates that the abnormal inventory data meets a first preset condition, the original image data, the structured inventory data, the abnormal reason and the evidence materials are summarized to generate an audit file. The application can improve the accuracy and efficiency of generating the audit file.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and more specifically to a data processing method, apparatus, and device. Background Technology

[0002] In related technologies, manual copying or general optical character recognition (OCR) is used to assist in inventory counting. In this process, operators photograph each physical device on-site, identify the device code, and then manually organize and enter the identified data into inventory forms to generate audit files.

[0003] However, manually compiling audit files is error-prone and inefficient. Summary of the Invention

[0004] The technical problem to be solved by this invention is that manual processing to generate audit files is prone to errors and inefficient.

[0005] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A data processing method, the method comprising: acquiring original image data corresponding to the physical terminal; determining structured inventory data based on the original image data and preset structured constraints; determining abnormal inventory data based on the structured inventory data and preset inventory benchmark data in response to a field change event; determining the abnormal cause and supporting materials corresponding to the abnormal inventory data based on the abnormal inventory data; performing compliance verification processing on the abnormal inventory data based on the abnormal cause and supporting materials to obtain a compliance verification processing result; if the compliance verification processing result indicates that the abnormal inventory data meets a first preset condition, summarizing the original image data, structured inventory data, abnormal cause and supporting materials to generate an audit file.

[0006] The beneficial effects of this invention are as follows: By acquiring the original image data corresponding to the physical terminal, determining structured inventory data based on the original image data and preset structured constraints, responding to the triggering of field change events, determining abnormal inventory data based on the structured inventory data and preset inventory benchmark data, determining the abnormal reasons and supporting materials corresponding to the abnormal inventory data, completing compliance verification processing based on the abnormal reasons and supporting materials to obtain compliance verification processing results, and generating audit files by summarizing the original image data, structured inventory data, abnormal reasons and supporting materials when the abnormal inventory data meets the first preset conditions, this fully automated processing method replaces the manual operation of manually sorting and identifying data, manually entering inventory forms and generating audit files in related technologies. This avoids the error probability caused by manual sorting operations and improves processing efficiency. At the same time, the processing logic of automatically judging abnormal inventory data based on structured inventory data and preset inventory benchmark data, verifying the abnormal reasons and supporting materials before archiving, can automatically identify abnormal inventory data and retain complete traceability information before generating audit files, solving the problems of not being able to automatically check inventory anomalies before archiving and lacking traceability basis for discrepancies in files.

[0007] Based on the above technical solution, the present invention can be further improved as follows.

[0008] Furthermore, in response to a field change event, based on structured inventory data and preset inventory benchmark data, abnormal inventory data is identified, including: writing the structured inventory data into a preset business form; if a field in the preset business form changes, performing regular expression validation on the device code within the structured inventory data to obtain a regular expression validation result; if the regular expression validation result indicates that the device code meets a second preset condition, checking whether preset inventory benchmark data matching the structured inventory data exists in the resource center; if preset inventory benchmark data matching the structured inventory data exists in the resource center, determining the confidence level corresponding to the structured inventory data based on the structured inventory data and the preset inventory benchmark data; if the confidence level is within the confidence level range or preset inventory benchmark data matching the structured inventory data does not exist in the resource center, identifying abnormal inventory data based on the structured inventory data.

[0009] The beneficial effect of adopting the above-mentioned further solution is that by using field changes as the trigger condition for automatic verification, invalid data is filtered out before verification using the device code regular expression format. Then, by combining the preset inventory benchmark data matching results and confidence level dual judgment rules in the resource center, abnormal inventory data is identified. This eliminates the need for manual verification of inventory data and inventory information item by item, avoids errors in human judgment, and thus improves the accuracy of audit files.

[0010] Furthermore, the preset business form includes cells, an inventory status field, and an inventory completion field. The method also includes: modifying the inventory status field corresponding to the abnormal inventory data so that the first inventory status corresponding to the inventory status field is changed to a difference pending confirmation status; modifying the inventory completion field corresponding to the abnormal inventory data so that the second inventory status corresponding to the abnormal inventory data is changed to a prohibited modification status; and adjusting the display style of the cells corresponding to the abnormal inventory data, configuring the cells to a highlighted display format.

[0011] The beneficial effect of adopting the above-mentioned further solution is that, through the pre-configured cells, inventory status fields, and inventory completion fields in the preset business form, after identifying abnormal inventory data, the inventory status field, inventory completion field, and cell corresponding to the abnormal inventory data are modified sequentially. The first inventory status corresponding to the inventory status field is changed to the difference pending confirmation status, the second inventory status corresponding to the inventory completion field is changed to the prohibition of modification status, and the display style of the cell corresponding to the abnormal inventory data is adjusted to configure the cell to a highlighted display format. In other words, through the automatic modification of the three types of objects in the preset business form, the form position corresponding to the abnormal inventory data is intuitively marked, the completion editing permission of the abnormal record is locked, and the abnormal status identifier is retained. There is no need to manually annotate the form and lock the items, avoiding the problems of manual annotation omissions and arbitrary tampering with the records corresponding to the abnormal inventory data.

[0012] Furthermore, based on the abnormal inventory data, the causes of the abnormality and supporting evidence are determined, including: sending abnormality alerts to the target object; in response to the target object's selection of the abnormality alert, determining the causes of the abnormality from the abnormality cause database; and in response to the target object's upload of the abnormality alert, determining the supporting evidence.

[0013] The beneficial effect of adopting the above-mentioned further solution is that through the linkage technology mechanism of pushing abnormal inventory data, determining the corresponding abnormal cause from the abnormal cause database based on the selection operation of abnormal reminder information, and determining the corresponding evidence materials based on the upload operation of abnormal reminder information, the target object can be reached immediately after abnormal inventory data is generated, thereby shortening the time delay of abnormal handling and avoiding the problem of abnormal omission.

[0014] Furthermore, based on the cause of the anomaly and the supporting evidence, compliance verification processing is performed on the abnormal inventory data to obtain the compliance verification results, including: detecting whether the supporting evidence and the cause of the anomaly match based on preset binding verification rules; wherein, the preset binding verification rules indicate the correlation between the target supporting evidence and the target cause of the anomaly; if the supporting evidence and the cause of the anomaly match, it indicates that the abnormal inventory data meets the first preset condition; if the supporting evidence and the cause of the anomaly do not match, it indicates that the abnormal inventory data does not meet the first preset condition.

[0015] The beneficial effect of adopting the above-mentioned further solution is that by using a preset binding verification rule to characterize the relationship between the target evidence materials and the target abnormal cause, it can detect whether the evidence materials and the abnormal cause match. Based on whether the evidence materials and the abnormal cause match, it can determine whether the compliance verification processing result meets the first preset condition, thereby realizing the automatic verification of evidence materials and abnormal causes and avoiding verification errors caused by inconsistent manual review standards.

[0016] Furthermore, the preset structured constraints include role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints. Based on the original image data and the preset structured constraints, the structured inventory data is determined, including: based on the original image data, role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints, the structured inventory data is determined using a preset neural network model.

[0017] The beneficial effect of adopting the above-mentioned further solution is that by inputting the original image data together with the preset structured constraints, which consist of role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints, into the preset neural network model to determine the structured inventory data, the inference output of the preset neural network model can be uniformly limited from multiple dimensions such as role recognition, output character, defect adaptation, sample reference, and data format. This overcomes the drawback of the messy and disordered output content in the unconstrained recognition mode, reduces the workload of subsequent manual image recognition content organization, and improves the accuracy and generation efficiency of structured inventory data.

[0018] Furthermore, the present invention provides a data processing apparatus, comprising: an acquisition module for acquiring original image data corresponding to the physical items at the terminal; a first determination module for determining structured inventory data based on the original image data and preset structured constraints; a second determination module for determining abnormal inventory data based on the structured inventory data and preset inventory baseline data in response to a field change event; a third determination module for determining the abnormal cause and supporting documentation corresponding to the abnormal inventory data; a verification module for performing compliance verification processing on the abnormal inventory data based on the abnormal cause and supporting documentation to obtain a compliance verification processing result; and a summary generation module for summarizing the original image data, structured inventory data, abnormal cause, and supporting documentation to generate an audit file if the compliance verification processing result indicates that the abnormal inventory data meets a first preset condition.

[0019] Furthermore, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the above-described data processing method by executing the computer instructions.

[0020] Furthermore, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the above-described data processing method.

[0021] Furthermore, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the above-described data processing method. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the structure of the data processing system provided by the present invention; Figure 2 A flowchart illustrating the data processing method provided by the present invention; Figure 3 A flowchart illustrating another data processing method provided by the present invention; Figure 4 A flowchart illustrating another data processing method provided by the present invention; Figure 5 A flowchart illustrating another data processing method provided by the present invention; Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0023] The principles and features of the present invention are described below. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0024] Currently, in the scenario of terminal inventory in telecommunications operator service halls, there are two main technical implementations: (1) Terminal physical inventory is carried out according to the inventory management system of the resource center. The inventory management system adopts the T+1 data synchronization mechanism every other day. The staff of the business hall collect the terminal serial number, i.e. IMEI data, by scanning the barcode or manually entering the data. The collected IMEI data cannot be compared with the inventory benchmark data in real time and the comparison results cannot be output immediately. Under this inventory method, the inventory data collection link, the inventory difference analysis link, and the abnormal control execution link are separated in time. The data collection and difference verification form an open-loop control link. The erroneous data generated by the inventory on the same day continues to accumulate before the synchronization on the next day and cannot be identified and intercepted in real time. When the difference verification is carried out after the data synchronization is completed on the next day, the on-site environment, physical placement and other on-site context information of the physical inventory have been lost. It is difficult to trace the real cause of the inventory difference. It is very easy to have the inventory difference not be effectively attributed, which ultimately causes the hidden danger of inventory loss control.

[0025] (2) The inventory is conducted using either manual copying or general optical character recognition (OCR) assisted inventory methods. In this method, operators photograph each terminal item on-site, identify the device code, and then manually organize the identification data and enter it into the inventory form to generate audit files. However, this manual process for generating audit files is prone to errors and inefficient.

[0026] Based on this, the present invention provides a data processing method, which involves acquiring the original image data corresponding to the physical terminal, determining structured inventory data based on the original image data and preset structured constraints, determining abnormal inventory data based on the structured inventory data and preset inventory baseline data in response to a field change event, determining the abnormal reasons and supporting materials corresponding to the abnormal inventory data, completing compliance verification processing based on the abnormal reasons and supporting materials to obtain the compliance verification processing result, and summarizing the original image data, structured inventory data, and abnormal reasons when the abnormal inventory data meets a first preset condition. This fully automated process for generating audit files from supporting evidence replaces the manual steps of data processing and identification, manual entry of inventory forms, and subsequent audit file generation in related technologies. This avoids the error probability caused by manual processing and improves processing efficiency. At the same time, it automatically identifies abnormal inventory data based on structured inventory data and preset inventory benchmark data, verifies the causes of abnormalities with supporting evidence, and then archives the data. It can automatically identify abnormal inventory data and retain complete traceability information before generating audit files, solving the problems of not being able to automatically check for inventory abnormalities before archiving and lacking traceability evidence for discrepancies in the files.

[0027] Figure 1 This is a schematic diagram of the data processing system provided by the present invention.

[0028] Combination Figure 1 As shown, the data processing system includes a perception layer module, a cognition layer module, a decision-making layer module, and an interaction layer module.

[0029] The perception layer module is used to acquire the original image data corresponding to the physical object on the terminal. Specifically, it can acquire the original image data by taking a picture with the mobile phone itself, overlay an immutable UTC timestamp on the original image data to complete the spatiotemporal anchoring of the original image, and then output the original image data carrying the anchoring information and send it to the cognition layer module.

[0030] The cognitive layer module has built-in preset structured constraints (role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints) and preset neural network models. The cognitive layer module receives the raw image data output by the perception layer, and based on the raw image data and preset structured constraints, uses the preset neural network model to generate structured inventory data, and pushes the structured inventory data to the decision layer module.

[0031] The decision-making module is used to listen for field change events caused by writing structured inventory data into a preset business form. After the field change event is triggered, the module performs regular expression format validation on the equipment code in the structured inventory data. After the validation passes, it retrieves the preset inventory baseline data in the resource center. Based on the baseline matching result and the calculated confidence level, it determines whether abnormal inventory data is generated. After abnormal inventory data is determined, it automatically modifies the status of the corresponding inventory status field, inventory completion field, and cell in the preset business form to achieve business-level circuit breaker processing. The interaction layer module pre-configures an anomaly cause database; it pushes anomaly alerts to operators via a Webhook interface; operators select an anomaly cause from the database and upload supporting documentation based on the alert; the interaction layer module then sends the selected anomaly cause and supporting documentation back to the corresponding verification link in the decision-making layer; based on preset binding verification rules, it performs compliance verification on the anomaly cause and supporting documentation; after successful verification, it aggregates all data in the entire link to generate an immutable audit file, and simultaneously unlocks form fields.

[0032] According to an embodiment of the present invention, a data processing method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] This embodiment provides a data processing method that can be used in the aforementioned data processing system. Figure 2 This is a flowchart of a data processing method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain the original image data corresponding to the physical terminal.

[0034] Raw image data can be understood as image data that carries the physical appearance of the terminal and the device identification information.

[0035] In this embodiment, the local camera can be used to take pictures of the terminal object (such as a mobile phone) to obtain the original image data.

[0036] Step S202: Based on the original image data and preset structured constraints, determine the structured inventory data.

[0037] Preset structured constraints can be understood as pre-defined constraints used to constrain the original image data in order to determine the constraints of the structured inventory data. These preset structured constraints may include, but are not limited to, role definition constraints, character constraints, imaging defect constraints, small sample constraints, and format output constraints.

[0038] Specifically, after determining the original image data, structured inventory data can be determined based on the original image data and preset structured constraints.

[0039] As an example, a neural network model can be used to determine structured inventory data based on raw image data and preset structured constraints. The raw image data and preset structured constraints serve as input to the neural network model, while the structured inventory data serves as its output. For instance, a sales office uploads multiple sets of raw images of equipment, some of which have issues such as label glare or corners obscured by tape. Using preset structured constraints, both the raw image data and the preset structured constraints are fed into the neural network model. Based on the imaging defect constraint, inference is performed to complete the missing IMEI characters. Small sample constraints retrieve 10 benchmark samples of normal and damaged labels from the system library to assist the model in comparison and recognition. Character constraints intercept non-numeric miscellaneous content, and format output constraints force the results to be encapsulated as a standard JSON message. After the model completes its calculations, JSON-formatted structured inventory data is generated line by line.

[0040] Step S203: In response to the field change event being triggered, abnormal inventory data is determined based on structured inventory data and preset inventory baseline data.

[0041] Field change events can be used to indicate the triggering conditions for initiating structured inventory data verification.

[0042] Preset inventory baseline data can be understood as standard reference data used to compare with inventory count data to determine whether the inventory count is abnormal.

[0043] Abnormal inventory data can be understood as data that is compliant in format but does not match the baseline data correctly.

[0044] Specifically, after determining the structured inventory data, abnormal inventory data can be identified based on the structured inventory data and preset inventory baseline data when a field change event is triggered.

[0045] As an example, a structured inventory count data entry carrying a 15-digit IMEI and a confidence level of 0.87 is written into a preset business form cell. The cell field content changes from blank to a specific code, triggering a field change event. The IMEI code is extracted and a 15-digit pure numeric regular expression format validation is performed, and the code fully complies with the specifications. The program connects to the resource center to retrieve the ledger. The resource center queries the preset inventory baseline data for the corresponding device. The inventory count data is compared with the baseline data, and the confidence level is calculated to be 0.87. This value falls within the confidence level range of 0.8 to 0.95, and is therefore determined to be abnormal inventory count data.

[0046] As an example, when structured inventory data is written to a form, triggering a field change event, the device IMEI is fully compliant with the regular expression format validation. After a full database traversal query, there is no preset inventory baseline data in the resource center ledger that matches the current IMEI. Without calculating the confidence level, the data is directly determined to be abnormal inventory data.

[0047] Step S204: Based on the abnormal inventory data, determine the abnormal causes and supporting evidence for the abnormal inventory data.

[0048] The cause of the anomaly can indicate the objective reason for the inventory discrepancy. Evidence can corroborate the authenticity and validity of the selected cause of the anomaly.

[0049] Specifically, the cause of the anomaly and supporting evidence can be determined after the abnormal inventory data is identified.

[0050] As an example, after receiving the abnormal inventory data, a corresponding abnormal reminder message is generated and pushed to the operator responsible for the inventory business. After receiving the reminder, the operator needs to enter the abnormal handling page within the preset shelf life, select the abnormal cause that matches the actual business scenario from the standardized options built into the abnormal cause library, and after selecting the abnormal cause, take photos of documents, physical objects and other materials on site and upload them to form evidence materials according to the material requirements corresponding to the abnormal cause.

[0051] Step S205: Based on the cause of the anomaly and the supporting evidence, perform compliance verification on the abnormal inventory data to obtain the compliance verification result.

[0052] The compliance verification process can be understood as the matching result between the feedback reasons and the evidence provided, which determines whether the data can enter the archiving process.

[0053] Specifically, the system loads preset binding verification rules, matches the correspondence between the detected anomaly cause and the supporting evidence, and if the two match and meet the rule requirements, the preset conditions are met and the verification is successful. If they do not match or the supporting evidence is missing, the verification fails, the abnormal data form is continuously locked, and entry into the archiving process is prohibited.

[0054] In one possible implementation, standard evidence material feature hash values ​​are pre-stored for each type of anomaly, forming binding verification rules. After a user uploads evidence materials, image features are extracted to generate local hash values, which are then compared with the standard template hash values. If the similarity meets the standard, the matching and verification are deemed successful; otherwise, the preset conditions are not met.

[0055] Step S206: If the compliance verification result indicates that the abnormal inventory data meets the first preset condition, summarize the original image data, structured inventory data, abnormal reasons and supporting materials, and generate an audit file.

[0056] Specifically, if the compliance verification result indicates that the abnormal inventory data meets the first preset condition, it means that the audit file can be generated. Then, the original image data, structured inventory data, reasons for the abnormality and supporting materials are summarized to generate the audit file.

[0057] As an example, the original image data, structured inventory data, reasons for anomalies, and supporting evidence are structurally assembled; at the same time, time stamps of each operation are embedded to form a time-series evidence chain, and tamper-proof file fingerprints are generated through encrypted hash algorithms, and standardized audit files are packaged together.

[0058] As an example, the archived content is divided into four main areas: raw image area, structured data area, anomaly handling area, and time log area, with each area storing its corresponding data. After the integrity of the data in each area is independently verified, it is uniformly packaged to generate a hierarchical audit archive.

[0059] The data processing method provided by this invention acquires the original image data corresponding to the physical items at the terminal, determines structured inventory data based on the original image data and preset structured constraints, determines abnormal inventory data in response to a field change event, determines the abnormal reasons and supporting materials based on the structured inventory data and preset inventory benchmark data, determines the abnormal reasons and supporting materials based on the abnormal inventory data, completes compliance verification processing based on the abnormal reasons and supporting materials to obtain the compliance verification results, and generates an audit file by summarizing the original image data, structured inventory data, abnormal reasons and supporting materials when the abnormal inventory data meets the first preset condition. This fully automated processing method replaces the manual operation of manually sorting and identifying data, manually entering inventory forms and generating audit files in related technologies, thereby avoiding the error probability caused by manual sorting operations and improving processing efficiency. At the same time, the processing logic of automatically judging abnormal inventory data based on structured inventory data and preset inventory benchmark data, verifying the abnormal reasons and supporting materials before archiving, can automatically identify abnormal inventory data and retain complete traceability information before generating audit files, solving the problems of not being able to automatically check inventory anomalies before archiving and lacking traceability basis for discrepancies in files.

[0060] Based on step S203, in response to the triggering of a field change event, and based on structured inventory data and preset inventory baseline data, the abnormal inventory data is determined to be processed as follows: Figure 3 The implementation of steps S3031 to S3035 in the above steps: Step S3031: Write the structured inventory data into a preset business form.

[0061] The pre-defined business forms can be understood as providing field monitoring entry points and data validation carriers for storing and displaying structured inventory data.

[0062] Specifically, the structured inventory data is written into the storage cell fields corresponding to the preset business form.

[0063] Step S3032: If the fields in the preset business form where the structured inventory data is written change, perform regular expression format validation on the equipment codes in the structured inventory data to obtain the regular expression format validation result.

[0064] The regular expression format validation result is used to indicate the format compliance status of the device encoding.

[0065] Preset conditions can be understood as pre-defined compliance standards for equipment coding formats, used to filter inventory data that are valid in format and can be used for benchmark comparison.

[0066] Specifically, the system monitors the data change status of preset business forms. When a structured inventory data write operation causes a change in the content of the corresponding field in the form, the verification mechanism is triggered to extract the device code from the structured inventory data, call the preset regular expression rules to complete the format screening, and generate the corresponding regular expression format verification result.

[0067] As an example, based on the backend resident listening thread monitoring the memory data of form fields, once a field change event caused by a write operation is captured, the device IMEI encoding string is immediately intercepted, the IMEI regular expression is loaded, and a full match verification is performed on the 15-digit numeric encoding format, the absence of special characters, and the encoding prefix rules. The regular expression format verification result is output, indicating whether the format is compliant or abnormal.

[0068] As an example, the form data change hook function captures field changes, triggers segmented regular expression format validation, verifies the encoding prefix, valid digits, and suffix validation bits step by step, filters out data with format defects, and generates regular expression format validation results.

[0069] Step S3033: If the regular expression format verification result indicates that the device code meets the second preset condition, check whether there is preset inventory benchmark data in the resource center that matches the structured inventory data.

[0070] The resource center can be used to store complete terminal device ledger data.

[0071] Preset inventory baseline data can be understood as standard terminal equipment ledger data stored in the resource center.

[0072] The second preset condition can be understood as the condition that the device encoding format is compliant. This second preset condition may include the absence of missing formatting or invalid data.

[0073] Specifically, the system reads the regular expression format verification result. Under the premise that the device encoding format is regular and meets the preset conditions, it actively initiates a data retrieval request to the resource center. Using the device IMEI code as the unique retrieval key, it iterates through the resource center to query whether there is preset inventory benchmark data that matches the current structured inventory data.

[0074] In one possible implementation, after the format verification is successful and the preset conditions are met, the IMEI index table pre-built in the resource center is called to quickly find the matching preset inventory baseline data through the index.

[0075] Step S3034: If the resource center has preset inventory benchmark data that matches the structured inventory data, determine the confidence level corresponding to the structured inventory data based on the structured inventory data and the preset inventory benchmark data.

[0076] Confidence level can characterize the similarity between structured inventory data and preset inventory benchmark data.

[0077] Specifically, after the resource center successfully retrieves the corresponding preset inventory baseline data, it analyzes the core information such as equipment parameters, coding features, and equipment attributes of the structured inventory data and the preset inventory baseline data. It then quantifies the degree of matching between the two sets of data through a similarity calculation algorithm and calculates the confidence level corresponding to the structured inventory data.

[0078] In one possible implementation, a hash feature comparison technique is used to generate feature hash values ​​for the structured inventory data and the preset inventory benchmark data respectively, and the corresponding confidence level is calculated and output through the hash difference.

[0079] Step S3035: If the confidence level is within the confidence level range or the resource center does not have preset inventory benchmark data that matches the structured inventory data, determine the abnormal inventory data based on the structured inventory data.

[0080] Specifically, a multi-condition fusion judgment logic is executed. On one hand, it determines whether the calculated confidence level falls within a preset confidence level range; on the other hand, it determines whether the resource center lacks matching preset inventory benchmark data. The step of determining whether the calculated confidence level falls within the preset confidence level range is only executed if the resource center does not lack matching preset inventory benchmark data. If either judgment condition is met, the current structured inventory data is determined to be abnormal inventory data. For example, if the confidence level falls within the 0.8~0.95 confidence level range, or if the resource center has no matching benchmark data, abnormal inventory data is directly generated if either condition is met.

[0081] If the confidence level is not within the confidence level range, determine whether the confidence level is greater than the maximum value of the confidence level range. If the confidence level is greater than the maximum value of the confidence level range, determine that the structured inventory data is normal inventory data.

[0082] If the confidence level is not within the confidence level range, determine whether the confidence level is less than the minimum value of the confidence level range. If the confidence level is less than the minimum value of the confidence level range, it is determined that the original image data has insufficient confidence level, the corresponding cell is locked, and the operator is prompted to re-terminate the image of the physical object, and no abnormal inventory data is generated.

[0083] The data processing method provided by this invention uses field changes as the trigger condition for automatic verification, filters invalid data before verification using device code regular expression format, and identifies abnormal inventory data by combining the matching results of preset inventory benchmark data in the resource center with the dual judgment rules of confidence level. This eliminates the need for manual verification of inventory data and inventory information item by item, avoids errors in human judgment, and thus improves the accuracy of audit files.

[0084] Based on step S204, which involves determining the cause of the anomaly based on the abnormal inventory data and specifying the supporting documentation, the appropriate method is as follows: Figure 4 The implementation of steps S4041 to S4043 is as follows: Step S4041: Send an anomaly alert message corresponding to the abnormal inventory data to the target object.

[0085] The target group can be understood as the operators responsible for checking and handling inventory anomalies.

[0086] Anomaly alerts can be understood as structured push notifications of anomalies, used to reach the target object and prompt the handling of inventory anomaly tasks.

[0087] Specifically, after identifying abnormal inventory data, the target object corresponding to this inventory operation can be located, and then an abnormal reminder message corresponding to the abnormal inventory data can be sent to the target object.

[0088] As an example, after generating abnormal inventory data, the system automatically encapsulates fields such as the abnormal device IMEI, abnormal status, generation time, and pending status, assembling them into standardized card-style abnormal alert messages. These messages are then linked to the on-duty warehouse manager's WeChat account via a Webhook interface, and the corresponding abnormal alert messages are sent to the on-duty warehouse manager.

[0089] Step S4042: In response to the target object's selection operation for the abnormal reminder information, determine the abnormal cause corresponding to the abnormal inventory data from the abnormal cause database.

[0090] Specifically, after the target user receives and clicks to process the exception alert, the system redirects to a standardized exception handling page, loading a pre-built exception cause library in the backend. Responding to the user's active selection, the system matches and identifies the exception cause corresponding to the current abnormal inventory data from the standardized tag set in the exception cause library. For example, after the target user clicks the alert card, they are automatically redirected to the system's exception handling page. The system loads the exception cause library, displaying standardized fixed options such as "sold but not billed," "goods in transit," "incorrect serial number," and "transferred but not returned to warehouse." Responding to the user's single selection, the system identifies the unique exception cause corresponding to the current abnormal inventory data.

[0091] Step S4043: In response to the target object's upload operation for the abnormal alert information, determine the supporting materials corresponding to the abnormal inventory data.

[0092] After selecting the corresponding cause of the anomaly for the target object, in response to the user's file upload operation, the system collects the documents, physical photos, circulation vouchers, and other materials uploaded by the user as needed, binds the uploaded voucher materials to the current anomaly inventory data, and generates unique corresponding evidence materials.

[0093] The data processing method provided by this invention, through a linkage mechanism of pushing abnormal inventory data with abnormal reminder information, determining the corresponding abnormal cause from the abnormal cause database based on the selection operation of the abnormal reminder information, and determining the corresponding evidence materials based on the upload operation of the abnormal reminder information, can reach the target object immediately after the abnormal inventory data is generated, shorten the time delay of abnormal handling and avoid the problem of abnormal omission.

[0094] Based on step S205, which involves verifying the abnormal inventory data according to the cause of the anomaly and supporting evidence, the compliance verification results are obtained as follows: Figure 5 The implementation of steps S5051 to S5053 is as follows: Step S5051: Based on the preset binding verification rules, detect whether the evidence materials and the cause of the anomaly match; wherein, the preset binding verification rules indicate the correlation between the target evidence materials and the target cause of the anomaly.

[0095] The preset binding verification rules can be understood as pre-configured and solidified association matching logic, which is used to constrain and bind the pairing relationship between various abnormal reasons and corresponding compliance evidence materials.

[0096] Specifically, after selecting the cause of the anomaly and uploading the supporting materials, the system automatically retrieves the pre-stored binding verification rules in the background. These rules pre-define the one-to-one correspondence between each type of target anomaly cause and the specific target supporting materials. Based on this fixed association logic, the system automatically compares and checks the anomaly cause actually submitted by the user with the supporting materials actually uploaded, and determines whether the material type, credential attributes, and content purpose of the two meet the standard matching requirements.

[0097] Step S5052: If the evidence materials and the cause of the anomaly match, it indicates that the abnormal inventory data meets the first preset condition.

[0098] When it is determined that the type and content of the evidence materials uploaded by the user are completely compatible with the currently selected cause of the anomaly and conform to the standard binding relationship, a compliance verification result with a qualified status is generated.

[0099] Step S5053: If the supporting materials and the cause of the anomaly do not match, it indicates that the abnormal inventory data does not meet the first preset condition.

[0100] If the system detects that the user-uploaded evidence is missing, the material type is incorrect, or the content of the voucher is irrelevant to the cause of the anomaly, the system determines that the two fail to match and generates a compliance verification result that is not in compliance.

[0101] The data processing method provided by this invention detects whether the evidence materials and the cause of the anomaly match by using preset binding verification rules to characterize the relationship between the target evidence materials and the target cause of the anomaly. Based on whether the evidence materials and the cause of the anomaly match, it determines whether the compliance verification processing result meets the preset conditions, thereby realizing automatic verification of the evidence materials and the cause of the anomaly and avoiding verification errors caused by inconsistent manual review standards.

[0102] In one possible implementation, the preset structured constraints include role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints. Step S202 above includes the following steps: Step a1: Based on the original image data, role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints, the structured inventory data is determined using a preset neural network model.

[0103] Role definition constraints can be used to limit the boundaries of the recognition tasks and roles of neural network models, eliminate invalid recognition and recognition of irrelevant content, and focus on the core task of device code recognition.

[0104] Character constraints can be used to standardize the character types, number of bits, and format ranges that the model can recognize, and to filter invalid characters and miscellaneous symbols.

[0105] Imaging defect constraints can be used to correct and infer defects such as reflection, occlusion, blur, tilt, and low light in real-world images, and to fix recognition deviations in non-ideal images.

[0106] Small sample constraints can be used to provide standard benchmark samples for reference, assisting the model in learning the characteristics of various device labels and improving the recognition and adaptation capabilities of niche, irregular, and incomplete label scenarios.

[0107] Output format constraints can be used to standardize the data structure, field specifications, and encapsulation format of the model output.

[0108] Specifically, the raw image data and five types of constraint parameters are synchronously input into a pre-defined neural network model, and the entire inference process of the model is standardized according to a multi-constraint collaborative management mechanism. Among these, the role definition constraint locks the model to only perform the device IMEI code recognition task, eliminating irrelevant interference information such as background and device appearance; the character constraint restricts the model to only recognize compliant numeric codes, blocking invalid content such as letters, symbols, and garbled characters; the imaging defect constraint addresses image defects such as label reflection, dirt obstruction, shooting tilt, and low-light blurring during actual shooting, performing feature repair and missing character inference completion; the small sample example constraint calls the system's built-in standard device label samples to assist the model in comparing and correcting feature deviations, adapting to various irregular, old, and damaged label recognition scenarios; and the format output constraint unifies the encapsulation structure and field format of the model results. Under the constraints of multiple constraints, the pre-defined neural network model completes image feature extraction, defect repair, character recognition, and similarity calibration, outputting structured inventory data with a unified format and complete information.

[0109] For example, in the daily intelligent inventory management scenario of a telecommunications store, staff take photos of the IMEI tags of terminal devices to generate raw image data carrying spatiotemporal information, initiating the intelligent recognition process. Five pre-configured constraint rules are retrieved, and the raw image data, along with role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints, are input into a pre-defined neural network model. Role definition constraints are used to lock in the device code recognition task, filtering out irrelevant background interference. Character constraints filter out 15-digit compliant numerical codes. For imaging issues such as tag reflections and slight occlusions caused by on-site shooting, pixel repair and character completion are performed based on imaging defect constraints. Standard benchmark samples from small sample example constraints are used to correct recognition deviations. Finally, format output constraints are used to organize the data structure. Regardless of whether the captured image is a standard, clear image or a non-standard image with slight imaging defects, the model can adaptively complete accurate inference, ultimately outputting structured inventory data containing complete IMEI codes and recognition confidence levels.

[0110] The data processing method provided by this invention inputs the original image data along with preset structured constraints, including role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints, into a preset neural network model to determine structured inventory data. This method can uniformly limit the inference output of the preset neural network model from multiple dimensions such as role recognition, output characters, defect adaptation, sample reference, and data format. It overcomes the drawbacks of messy and disordered output content in the unconstrained recognition mode, reduces the workload of subsequent manual image recognition content organization, and improves the accuracy and generation efficiency of structured inventory data.

[0111] In one possible implementation, the pre-defined business form includes cells, an inventory status field, and an inventory completion field. The above method may also include the following steps: Step b1: Modify the inventory status field corresponding to the abnormal inventory data so that the first inventory status corresponding to the inventory status field is changed to the difference pending confirmation status.

[0112] The inventory status field can be understood as a business field in a business form used to record the verification progress and difference attributes of a single inventory record.

[0113] The first inventory status can be understood as the original default status corresponding to the inventory status field, which can be a normal, verified, or no difference, etc.

[0114] The "Difference Pending Confirmation" status can be used to identify business statuses where there are discrepancies in the current inventory data and the cause needs to be manually verified and attributed.

[0115] Specifically, after determining that abnormal inventory data has been generated, the inventory status field corresponding to the data in the preset business form is located, and the first inventory status of the inventory status field is overwritten and modified, changing the original normal and undifferentiated default status to a difference pending confirmation status.

[0116] Step b2: Modify the inventory completion field corresponding to the abnormal inventory data so that the second inventory status corresponding to the abnormal inventory data is changed to a prohibited modification status.

[0117] The "Inventory Completion" field can be understood as an access control field in a business form that manages the editing permissions and locked status of a single inventory record.

[0118] The second inventory status can be understood as the original default status corresponding to the inventory completion field, which can be an editable, modifiable, or unlocked open permission status.

[0119] The "Do Not Modify" status can be used to prevent manual alteration or correction of abnormal inventory data.

[0120] Specifically, for the identified abnormal inventory data, the inventory completion field of the corresponding form is locked, and the permission of the original open and editable second inventory status of the field is switched and modified, forcibly changing the field permission status to a prohibited modification status.

[0121] Step b3: Adjust the display style of the cells corresponding to the abnormal inventory data, and configure the cells to be highlighted.

[0122] Display style can be understood as the original default display style of the cell, specifically the standard display style with no highlight and default background color.

[0123] Highlighting formatting can indicate the visual display style of cell anomalies.

[0124] Specifically, locate the corresponding display cell in the business form for the abnormal inventory data, modify the default display style of the cell, and switch the display style to a highlighted format.

[0125] In one scenario, during routine fully automated real-time inventory checks in a telecommunications service hall, the inventory status field of this record is modified. The default first inventory status (normal and without discrepancies) is updated to a discrepancy pending confirmation status, marking that this inventory data has an inventory discrepancy and requires manual verification and attribution. Simultaneously, the inventory completion field of this record is locked, and the editable second inventory status is switched to a prohibited modification status to prevent operators from privately modifying or erasing inventory discrepancy data during the anomaly verification phase. At the same time, the form cell corresponding to this data is switched to a highlighted red format.

[0126] The data processing method provided by this invention, through pre-configured cells, inventory status fields, and inventory completion fields in a preset business form, sequentially modifies the inventory status field, inventory completion field, and cell corresponding to the abnormal inventory data after identifying abnormal inventory data. This involves changing the first inventory status corresponding to the inventory status field to a "difference pending confirmation" status, changing the second inventory status corresponding to the inventory completion field to a "modification prohibited" status, and adjusting the display style of the cell corresponding to the abnormal inventory data, configuring the cell to a highlighted display format. In other words, by automatically modifying these three types of objects within the preset business form, the method intuitively marks the form position corresponding to the abnormal inventory data, locks the completion editing permissions of the abnormal records, and retains the abnormal status identifier. This eliminates the need for manual annotation of forms and locking of entries, avoiding the problems of manual omissions and arbitrary tampering with the records corresponding to the abnormal inventory data.

[0127] In some embodiments, the data processing apparatus of the present invention can be implemented in a combination of hardware and software. As an example, the data processing apparatus of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the data processing method of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0128] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0129] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-described data processing methods. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the data processing method shown in any embodiment of the present invention by calling the computer program.

[0130] In one alternative embodiment, an electronic device is provided, such as Figure 6 As shown, Figure 6 The illustrated electronic device 6000 includes a processor 6001 and a memory 6003. The processor 6001 and the memory 6003 are connected, for example, via a bus 6002. Optionally, the electronic device 6000 may further include a transceiver 6004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 6004 is not limited to one type, and the structure of the electronic device 6000 does not constitute a limitation on the embodiments of the present invention.

[0131] Processor 6001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 6001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0132] Bus 6002 may include a path for transmitting information between the aforementioned components. Bus 6002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 6002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The bus 6002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0133] The memory 6003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0134] The memory 6003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 6001. The processor 6001 executes the application code stored in the memory 6003 to implement the content shown in the foregoing method embodiments.

[0135] Among them, electronic devices can also be terminal devices, which can be any device that can install applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0136] It should be noted that, Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0137] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-described data processing methods.

[0138] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0139] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the data processing method described above.

[0140] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0141] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0142] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EEPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0143] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0144] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0145] It should be noted that the terms "first," "second," etc., used in the specification and claims of this invention are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of the invention described herein can be implemented in an order other than that shown or described.

[0146] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a circuit, module, or system. Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0147] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A data processing method, characterized in that, The method includes: Acquire the original image data corresponding to the physical terminal; Based on the original image data and preset structured constraints, structured inventory data is determined; In response to a field change event being triggered, abnormal inventory data is determined based on the structured inventory data and preset inventory baseline data; Based on the abnormal inventory data, determine the abnormal causes and supporting evidence for the abnormal inventory data. Based on the aforementioned causes of the anomaly and the supporting evidence, the abnormal inventory data is subjected to compliance verification processing to obtain the compliance verification processing result. If the compliance verification result indicates that the abnormal inventory data meets the first preset condition, the original image data, structured inventory data, reasons for the abnormality and supporting materials are summarized to generate an audit file.

2. The data processing method according to claim 1, characterized in that, In response to a field change event being triggered, based on the structured inventory data and preset inventory baseline data, abnormal inventory data is identified, including: Write the structured inventory data into a preset business form; If the fields in the preset business form where the structured inventory data is written change, the device codes in the structured inventory data are validated using regular expressions to obtain the regular expression validation result. If the regular expression format verification result indicates that the device code meets the second preset condition, check whether there is preset inventory benchmark data in the resource center that matches the structured inventory data; If the resource center has preset inventory benchmark data that matches the structured inventory data, the confidence level corresponding to the structured inventory data is determined based on the structured inventory data and the preset inventory benchmark data. If the confidence level is within the confidence level range or the resource center does not have preset inventory benchmark data that matches the structured inventory data, abnormal inventory data is determined based on the structured inventory data.

3. The data processing method according to claim 2, characterized in that, The pre-defined business form includes cells, an inventory status field, and an inventory completion field. The method also includes: Modify the inventory status field corresponding to the abnormal inventory data so that the first inventory status corresponding to the inventory status field is changed to a difference pending confirmation status. Modify the inventory completion field corresponding to the abnormal inventory data so that the second inventory status corresponding to the abnormal inventory data is changed to a prohibited modification status. Adjust the display style of the cells corresponding to the abnormal inventory data, and configure the cells to be highlighted.

4. The data processing method according to claim 1, characterized in that, Based on the abnormal inventory data, determine the cause of the abnormality and the supporting evidence, including: Send the abnormal inventory data corresponding to the target object with an abnormal alert message; In response to the target object's selection operation for the abnormal reminder information, the abnormal cause corresponding to the abnormal inventory data is determined from the abnormal cause database; In response to the target object's upload operation for the abnormal alert information, the supporting evidence materials corresponding to the abnormal inventory data are determined.

5. The data processing method according to claim 1, characterized in that, Based on the stated causes of the anomalies and the supporting evidence, the abnormal inventory data undergoes compliance verification processing to obtain the compliance verification results, including: Based on preset binding verification rules, it is detected whether the evidence materials and the cause of the anomaly match; wherein, the preset binding verification rules indicate the correlation between the target evidence materials and the target cause of the anomaly. If the evidence and the cause of the anomaly match, it indicates that the abnormal inventory data meets the first preset condition; If the evidence provided does not match the cause of the anomaly, it indicates that the abnormal inventory data does not meet the first preset condition.

6. The data processing method according to claim 1, characterized in that, The preset structured constraints include role definition constraints, character constraints, imaging defect constraints, small sample example constraints, and format output constraints. Based on the original image data and the preset structured constraints, the structured inventory data is determined, including: Based on the original image data, the role definition constraints, the character constraints, the imaging defect constraints, the small sample constraints, and the format output constraints, the structured inventory data is determined using a preset neural network model.

7. A data processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire the original image data corresponding to the physical terminal object; The first determining module is used to determine the structured inventory data based on the original image data and preset structured constraints; The second determination module is used to determine abnormal inventory data based on the structured inventory data and preset inventory baseline data in response to the triggering of a field change event. The third determining module is used to determine the cause of the abnormality and supporting evidence based on the abnormal inventory data. The verification module is used to perform compliance verification processing on the abnormal inventory data based on the cause of the abnormality and the supporting materials, and obtain the compliance verification processing result. The summary generation module is used to generate an audit file by summarizing the original image data, structured inventory data, reasons for the anomaly, and supporting materials if the compliance verification processing result indicates that the abnormal inventory data meets the first preset conditions.

8. An electronic device, characterized in that, include: A processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to cause the processor to perform the data processing method as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the data processing method according to any one of claims 1-6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the data processing method according to any one of claims 1-6.

Images