Paper work structured collection and tracing method and system based on standardized dynamic code point

Abstract

The application relates to the technical field of big data mining, and discloses a paper work structured collection and tracing method and system based on standardized dynamic code points, which comprises a dynamic code point generation module, a work collection module, a structured processing module, a tracing analysis module, a visual feedback module and a self-adaptive optimization module; the method realizes unique identification and metadata binding of paper work based on standardized dynamic code points, intelligently extracts work text content by combining image preprocessing and optical character recognition technology, and generates structured work data through natural language processing technology; dynamic code point query and tracing analysis algorithm are used to realize tracking and visual report output of work processing history and modification records. The application guarantees accurate association of work identity information and anti-fake collection data, further improves the reliability of collection, guarantees the accuracy of text content, and guarantees the integrity of tracing query and the reliability of visual report.

Description

Technical Field

[0001] This invention relates to the field of big data mining technology, specifically to a method and system for structured collection and traceability of paper-based operations based on standardized dynamic code points. Background Technology

[0002] Big data mining is a data processing technique that extracts hidden information from large amounts of data. Its core process includes data preprocessing, mining implementation, and pattern evaluation, and it is applied in fields such as business analysis and pattern recognition.

[0003] Currently, due to various environmental interferences and diverse work formats during the structured collection of paper-based work, the image acquisition equipment cannot adjust the lighting and angle parameters in real time when collecting images of paper-based work. This can lead to errors in the recognition of standardized dynamic code points. When the dynamic code point decoding fails, it can cause errors in the association of work identity information, making it impossible to guarantee the uniqueness and anti-counterfeiting of the collected data.

[0004] Therefore, a method and system for structured data collection and traceability of paper-based operations based on standardized dynamic code points are proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for structured acquisition and traceability of paper-based work based on standardized dynamic code points. This solves the problems mentioned in the background art, such as the inability of the image acquisition equipment to adjust lighting and angle parameters in real time when acquiring images of paper-based work, and the inability to guarantee the uniqueness and anti-counterfeiting properties of the acquired data.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method and system for structured data collection and traceability of paper-based operations based on standardized dynamic code points, wherein the method includes the following steps: S1. Generate standardized dynamic code points using a standardized dynamic code point generation device. The standardized dynamic code points contain a unique identifier and are associated with the metadata of the paper work, thereby generating dynamic code point data. S2. Use an image acquisition device to scan the paper work with the standardized dynamic code points attached, obtain work image data, and use a preprocessing unit to perform image enhancement and noise reduction on the work image data to generate preprocessed image data. S3. Based on the dynamic code point data, the identity information of the paper job is parsed, and the job text content is extracted from the preprocessed image data through the optical character recognition unit to generate the original text data; S4. Natural language processing technology is used to perform structured parsing on the original text data, identify the structural elements of the assignment's questions, answers, and scoring areas, and generate structured assignment data. S5. Bind the structured job data with the dynamic code point data, and save it to the database through the data storage unit to establish a job data index; S6. Based on the dynamic code point data, query the processing history, modification records and source information of the job, and generate a job traceability report using the traceability analysis algorithm; S7. Output the job traceability report in chart form through the visualization unit, and provide an interactive query interface for users to access.

[0007] Preferably, in step S1, standardized dynamic code points are generated using a standardized dynamic code point generation device. Each standardized dynamic code point contains a unique identifier and is associated with metadata from the paper-based work. Generating dynamic code point data includes the following steps: S11. Generate a unique identifier using the standardized dynamic code point generation device and associate it with the metadata of the paper work to generate the dynamic code point data; S12. Using a printing device, the standardized dynamic code dots are attached to a preset position on the paper work, and the size, density, and visibility parameters of the code dots are set through a configuration unit. S13. The dynamic code point data is bound to the metadata of the job creator, creation time, and course information through the metadata association unit to generate code point association data; The dynamic code point generation process is represented by the following formula: ; in, For dynamic code point data, As a unique identifier, For timestamps, As a random factor, This is a generation function based on an encryption algorithm.

[0008] Preferably, in step S2, the process of scanning paper worksheets with the standardized dynamic code dots using an image acquisition device to obtain worksheet image data, and then performing image enhancement and noise reduction on the worksheet image data through a preprocessing unit to generate preprocessed image data includes the following steps: S21. Use the image acquisition device to scan the paper work with the standardized dynamic code points attached, obtain the work image data, and adjust the image brightness and contrast through the light compensation unit; S22. The preprocessing unit performs grayscale conversion, binarization, and edge detection on the work image data to remove background noise and distortion. S23. The standardized dynamic code points are located and decoded from the operation image data by the code point recognition unit, the validity of the code points is verified, and the verification result is output to the preprocessed image data.

[0009] Preferably, step S3, which involves parsing the identity information of the paper job based on the dynamic code point data and extracting the job text content from the preprocessed image data using an optical character recognition unit to generate the original text data, includes the following steps: S31. Based on the dynamic code point data, retrieve the associated job metadata from the database and load the job template information; S32. The text region in the preprocessed image data is extracted by the optical character recognition unit, and character segmentation and recognition are performed to generate the original text data. S33. Use the grammar correction unit to perform spell checking and format standardization on the original text data.

[0010] Preferably, step S4 involves using natural language processing technology to perform structured parsing of the original text data, identifying the structural elements of the assignment's questions, answers, and grading areas, and generating structured assignment data, including the following steps: S41. Use a deep learning model to perform entity recognition on the original text data to distinguish between question, answer, and annotation structure blocks; S42. The logical structure of the job is parsed through the rule engine to generate structured JSON data as the structured job data; S43. Use the data verification unit to check the integrity and consistency of the structured operation data, and automatically repair missing and conflicting content; The structured parsing process is represented by the following formula: ; in, For structured operation data, The original text data, To parse parameters including the recognition thresholds for questions and answers, For deep learning models, This is a parsing function for natural language processing.

[0011] Preferably, in step S5, binding the structured job data with the dynamic code point data and saving it to the database through the data storage unit to establish a job data index includes the following steps: S51. The structured job data and the dynamic code point data are bound together by a mapping unit, and the data is securely encoded by an encryption unit. S52. Utilize a distributed storage system to save the bound data and establish a fast query index based on dynamic code points; S53. The modification history of the structured job data is recorded through the version control unit to ensure data traceability.

[0012] Preferably, step S6, which involves querying the processing history, modification records, and source information of a job based on the dynamic code point data, and generating a job source tracing report using a source tracing analysis algorithm, includes the following steps: S61. Based on the dynamic code point data, query the job processing records in the database, including the collection time, processing personnel, and modification operations; S62. Use blockchain technology to write key operation records into an immutable distributed ledger; S63. The data analysis unit statistically analyzes the workflow and efficiency of the operation and generates analysis charts for the operation traceability report.

[0013] Preferably, step S7, which outputs the job traceability report in chart form through a visualization unit and provides an interactive query interface for user access, includes the following steps: S71. The operation traceability report is converted into line chart, heat map and time axis format through the visualization unit; S72. Provide a user query interface using web interfaces and mobile applications, supporting data filtering by code point, time and personnel; S73. Real-time notification of abnormal tracing events is provided through the alarm unit.

[0014] Preferably, the system includes a dynamic code point generation module, a job acquisition module, a structured processing module, a traceability analysis module, a visualization feedback module, and an adaptive optimization module; The dynamic code point generation module generates unique dynamic code points using standardized coding units, associates job information through metadata binding units, and outputs code point data through the print control unit. The job acquisition module receives the code point data, acquires the job image through the image acquisition unit, optimizes the image using the preprocessing unit, and verifies the validity of the code points through the code point recognition unit. The structured processing module receives the job image, extracts the text content through the character recognition unit, parses the job structure using the natural language processing unit, and saves the structured data through the data storage unit. The traceability analysis module receives the structured data, retrieves the operation history through the query unit, ensures the data is tamper-proof through the blockchain record unit, and generates a traceability report through the analysis unit. The visualization feedback module receives the source tracing report, creates visualization output through the chart generation unit, provides user interaction through the interface unit, and sends notifications through the alarm unit. The adaptive optimization module receives the output of the source analysis module, evaluates the system efficiency through the performance monitoring unit, optimizes the processing parameters through the parameter adjustment unit, and improves the recognition algorithm through the learning unit.

[0015] Preferably, the dynamic code point generation module further includes a code point update unit to periodically refresh the dynamic code points to enhance security; The task acquisition module supports collaborative acquisition by multiple devices and enables real-time data upload through a cloud synchronization unit. The structured processing module integrates multiple OCR engines and adaptively selects the optimal algorithm; The source tracing and analysis module supports cross-platform data integration and provides API interfaces for third-party systems to call. The visual feedback module allows for customized report templates and supports multilingual output; The adaptive optimization module dynamically adjusts system parameters based on user feedback.

[0016] Compared with existing technologies, this invention provides a method and system for structured data collection and traceability of paper-based work based on standardized dynamic code points, which has the following advantages: 1. In this invention, when performing structured collection and traceability of paper-based operations, a unique identifier is generated by a standardized dynamic code generation device and associated with the metadata of the paper-based operations. At the same time, a preprocessing unit is used to perform real-time image enhancement and noise reduction on the operation image data acquired by the image acquisition device. This can solve the problem of standardized dynamic code recognition errors caused by ambient light and angle interference, ensure the accurate association of operation identity information and the anti-counterfeiting of the collected data, and further improve the reliability of collection.

[0017] 2. In this invention, when performing structured collection and traceability of paper-based tasks, the task text content is extracted from the preprocessed image data by the optical character recognition unit, and combined with the grammar correction unit for real-time spell checking and format standardization. This can detect and correct the recognition errors of handwritten and printed variants in real time, avoid errors in the extraction of original text data, ensure the accuracy of the text content, and thus improve the correctness of subsequent processing.

[0018] 3. In this invention, when performing structured collection and tracing of paper-based assignments, natural language processing technology is used to perform structured parsing of the original text data. A deep learning model is used to identify structural elements such as assignment questions, answers, and scoring areas. Combined with a rule engine, multimodal layout is adaptively matched to address the problem of inconsistent assignment partitioning, solve the problem of logical deviation in the generation of structured assignment data, ensure the integrity of tracing queries and the credibility of visualization reports, and enhance the applicability of the system. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the architecture of the paper-based work structured collection and traceability system based on standardized dynamic code points according to the present invention; Figure 2This is a flowchart illustrating the steps of the paper-based structured data collection and traceability method based on standardized dynamic code points, as described in this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1-2 The specific implementation of the paper-based structured data collection and traceability method and system based on standardized dynamic code points is as follows, and the method includes the following steps: S1. Generate standardized dynamic code points using a standardized dynamic code point generation device. Each standardized dynamic code point contains a unique identifier and is associated with the metadata of the paper-based work, thus generating dynamic code point data. S2. Use an image acquisition device to scan paper work with standardized dynamic code points to obtain work image data, and use a preprocessing unit to perform image enhancement and noise reduction on the work image data to generate preprocessed image data. S3. Based on dynamic code point data, analyze the identity information of paper work and extract the work text content from the preprocessed image data through the optical character recognition unit to generate the original text data; S4. Use natural language processing technology to perform structured parsing on the original text data, identify the structural elements of the assignment's questions, answers, and grading areas, and generate structured assignment data. S5. Bind the structured job data with the dynamic code point data, and save it to the database through the data storage unit to establish a job data index; S6. Based on dynamic code point data, query the processing history, modification records and source information of the job, and generate a job traceability report using the traceability analysis algorithm; S7. Output the job traceability report in chart form through visualization units and provide an interactive query interface for users to access.

[0022] In S1, standardized dynamic code points are generated using a standardized dynamic code point generation device. Each standardized dynamic code point contains a unique identifier and is associated with metadata from the paper-based work. The generation of dynamic code point data includes the following steps: S11. Generate a unique identifier using a standardized dynamic code point generation device and associate it with the metadata of the paper work to generate dynamic code point data; S12. Use a printing device to attach standardized dynamic code dots to a preset position on a paper workpiece, and set the size, density, and visibility parameters of the code dots through a configuration unit. S13. By binding dynamic code point data with the metadata of the job creator, creation time, and course information, code point association data is generated through the metadata association unit. The dynamic code point generation process is represented by the following formula: ; in, For dynamic code point data, As a unique identifier, For timestamps, As a random factor, This is a generation function based on an encryption algorithm; Specifically, generating functions Using the national standard SM4 encryption algorithm, firstly... The data is concatenated and then encrypted using the SM4 algorithm with a preset 128-bit key. The output is dynamic code point data. .

[0023] In S2, an image acquisition device scans paper-based worksheets with standardized dynamic code dots to obtain worksheet image data. The preprocessing unit then performs image enhancement and noise reduction on the worksheet image data to generate preprocessed image data, including the following steps: S21. Use an image acquisition device to scan paper work with standardized dynamic code dots to obtain work image data, and adjust the image brightness and contrast through a light compensation unit. S22. The preprocessing unit performs grayscale conversion, binarization, and edge detection on the work image data to remove background noise and distortion. S23. The code point recognition unit locates and decodes standardized dynamic code points from the work image data, verifies the validity of the code points, and outputs the verification result to the preprocessed image data. The image preprocessing process is represented by the following formula, and includes specific operational steps: ; in, For preprocessed image data, For the image data of the operation, For image enhancement parameters, For noise reduction parameters, This is an image preprocessing function; The specific operation steps include: performing brightness equalization on the job image data using an adaptive thresholding algorithm, then using a median filtering algorithm to remove salt-and-pepper noise, and finally outputting the preprocessed image data, image enhancement parameters, etc. To limit the contrast of the adaptive histogram equalization cropping, the value range is [2.0, 3.0]. Noise reduction parameters... This is the kernel size for median filtering, with a value of 5x5 pixels.

[0024] S3 parses the identity information of paper assignments based on dynamic code point data, and extracts the assignment text content from the preprocessed image data using the optical character recognition unit to generate the original text data, including the following steps: S31. Retrieve associated job metadata from the database based on dynamic code point data, and load job template information; S32. The text region in the preprocessed image data is extracted by the optical character recognition unit, and character segmentation and recognition are performed to generate the original text data. S33. Use the grammar correction unit to perform spell checking and format standardization on the original text data to improve text accuracy; The optical character recognition process is represented by the following formula, and includes specific operational steps: ; in, The original text data, For preprocessed image data, For OCR engine parameters, For optical character recognition functions; The specific operation steps include: first, segmenting text lines using a connected component analysis algorithm; then, using a convolutional neural network model to identify individual characters; and finally, correcting recognition errors using a dictionary matching algorithm to generate the original text data. The convolutional neural network model adopts a model based on a convolutional recurrent neural network structure, which includes 8 convolutional layers for feature extraction, 2 bidirectional long short-term memory networks for sequence modeling, and a connected temporal classification output layer for character recognition. The model is trained using a dataset containing Chinese and English characters and mathematical symbols.

[0025] S4 uses natural language processing technology to perform structured parsing of the raw text data, identifying the structural elements of the assignment's questions, answers, and grading areas, and generating structured assignment data, including the following steps: S41. Use a deep learning model to perform entity recognition on the original text data to distinguish between question, answer, and annotation structure blocks; S42. The logical structure of the job is parsed through the rule engine to generate structured JSON data as structured job data. JSON is a lightweight data exchange format. Structured JSON data is the standard format and carrier for the system to process and exchange "structured job data". S43. Use the data verification unit to check the integrity and consistency of structured operation data and automatically repair missing and conflicting content; The structured parsing process is represented by the following formula: ; in, For structured operation data, The original text data, To parse parameters including the recognition thresholds for questions and answers, For deep learning models, Parsing functions for natural language processing, deep learning models A BERT-based sequence labeling model was adopted and fine-tuned to enable it to recognize tags such as "start of question", "question content", "start of answer", and "answer content" and parse parameters. The identification threshold is set to 0.85, meaning that only labels with a model prediction probability higher than this threshold are adopted.

[0026] In S5, structured job data is bound to dynamic code point data and saved to the database through a data storage unit. The steps to establish a job data index are as follows: S51. Structured operation data and dynamic code point data are bound together through a mapping unit, and the data is securely encoded through an encryption unit. S52. Utilize a distributed storage system to save the bound data and establish a fast query index based on dynamic code points; S53. Record the modification history of structured job data through the version control unit to ensure data traceability; The data binding and index creation process is represented by the following formula, which includes specific operational steps: ; in, Indexing job data For structured operation data, For dynamic code point data, Functions for generating indexes; The specific operational steps include: first, mapping structured job data and dynamic code point data into key-value pairs using a hash algorithm; then, constructing an index using a B-tree structure; and finally, distributing the data across multiple nodes using a consistent hash algorithm to ensure query efficiency. The hash algorithm uses the SHA-256 algorithm for the dynamic code point data. Perform a hash operation, and use the resulting hash value as the key to structure the job data. As a value, the order of the B-tree index is set to 512 to balance query performance and storage overhead.

[0027] In S6, the processing history, modification records, and source information of a job are queried based on dynamic code point data. The process of generating a job source tracing report using a source tracing analysis algorithm includes the following steps: S61. Query the job processing records in the database based on dynamic code point data, including collection time, processing personnel and modification operations; S62. Use blockchain technology to write key operation records into an immutable distributed ledger to enhance traceability credibility; S63. Through the data analysis unit, the workflow path and efficiency of the operation are statistically analyzed, and analytical charts for the operation traceability report are generated. The source tracing analysis process is represented by the following formula, and includes specific operational steps: ; in, For the work traceability report, For dynamic code point data, To process historical data, For source tracing analysis functions; The specific operation steps include: first, extracting key event points from the processing history using time series analysis algorithms; then, constructing a flow path model using a graph database; and finally, generating a source tracing report using an aggregation algorithm, including anomaly detection and trend analysis. The time series analysis algorithm uses an outlier detection algorithm based on a density-based clustering method with noise to identify key events such as abnormal processing time intervals. The graph database uses Neo4j, where nodes are defined as "jobs", "personnel", and "operations", and relationships are defined as "executed" and "modified".

[0028] In S7, the job traceability report is output in chart form through visualization units, and an interactive query interface is provided for users to access it, including the following steps: S71. Convert the job traceability report into line charts, heatmaps, and timelines using visualization units; S72. Provide a user query interface using web interfaces and mobile applications, supporting data filtering by code point, time and personnel; S73. Real-time notification of abnormal tracing events through the alarm unit; The visualization output process is represented by the following formula, and includes specific operation steps: ; in, For visualization output, For the work traceability report, Visualization parameters include chart type and interaction settings. For visualization functions; The specific steps include: First, mapping the source tracing report data into scalable vector graphics using the D3.js library, a JavaScript library based on web standards primarily for data visualization; then, using responsive design to adapt to different device screens; and finally, using the WebSocket protocol to achieve real-time data updates and user interaction, and visualization parameters. In this framework, the mapping rules between chart types and data attributes are defined as follows: a line chart is used to represent the time span of job processing, a heat map is used to compare the processing time of each stage, a timeline is used for the complete processing chain, and interactive settings include clicking on chart elements to drill down and view detailed operation logs.

[0029] The system includes a dynamic code point generation module, a job acquisition module, a structured processing module, a source analysis module, a visualization feedback module, and an adaptive optimization module; The dynamic code point generation module uses standardized coding units to generate unique dynamic code points, associates job information through metadata binding units, and outputs code point data through the print control unit. The job acquisition module receives code point data, acquires job images through the image acquisition unit, optimizes the images using the preprocessing unit, and verifies the validity of the code points through the code point recognition unit. The structured processing module receives the job image, extracts the text content through the character recognition unit, parses the job structure using the natural language processing unit, and saves the structured data through the data storage unit. The traceability analysis module receives structured data, retrieves job history through the query unit, uses the blockchain record unit to ensure that the data is tamper-proof, and generates a traceability report through the analysis unit. The visualization feedback module receives the source tracing report, creates visualization output through the chart generation unit, provides user interaction through the interface unit, and sends notifications through the alarm unit. The adaptive optimization module receives the output from the source analysis module, evaluates the system efficiency through the performance monitoring unit, optimizes the processing parameters through the parameter adjustment unit, and improves the recognition algorithm through the learning unit.

[0030] The dynamic code point generation module also includes a code point update unit, which periodically refreshes the dynamic code points to enhance security; The job acquisition module supports collaborative acquisition by multiple devices and enables real-time data upload through the cloud synchronization unit; The structured processing module integrates multiple OCR engines and adaptively selects the optimal algorithm; The source tracing and analysis module supports cross-platform data integration and provides API interfaces for third-party systems to call; The visual feedback module allows for custom report templates and supports multilingual output; The adaptive optimization module dynamically adjusts system parameters based on user feedback, improving data acquisition accuracy and tracing efficiency.

[0031] The operational steps of the paper-based structured data collection and traceability method and system based on standardized dynamic code points are as follows: Step 1: Standardized Dynamic Code Point Generation: A unique identifier is generated by a standardized dynamic code generation device, and associated with the metadata of the paper work, including the creator, creation time, and course information, to generate dynamic code data. This step includes using an encoding unit to ensure the uniqueness and anti-counterfeiting of the code, attaching the dynamic code to a preset position on the paper work through a printing device, configuring the code size, density, and visibility parameters, and finally binding the work information through a metadata association unit to provide a basic identifier for subsequent collection and traceability.

[0032] Step 2: Image Acquisition and Preprocessing Using image acquisition devices, including high-resolution scanners and smartphone cameras, paper worksheets with standardized dynamic code dots are scanned to obtain worksheet image data. The image is then enhanced and denoised by a preprocessing unit. Specifically, the brightness and contrast of the image are adjusted by a light compensation unit, and grayscale, binarization, and edge detection algorithms are used to remove background noise and distortion to generate preprocessed image data, ensuring that the image quality meets the requirements of subsequent optical character recognition.

[0033] Step 3: Identity Information Parsing and Text Extraction The process involves parsing the identity information of paper-based assignments based on dynamic code point data, extracting the assignment text content from preprocessed image data using an optical character recognition unit, and generating original text data. This step includes retrieving associated assignment metadata from the database, loading assignment template information, extracting text regions using character segmentation and recognition algorithms, and performing spell checking and format standardization using a grammar correction unit to improve text accuracy.

[0034] Step 4: Structured parsing of job data: Natural language processing technology is used to perform structured parsing of raw text data, identify structural elements such as assignment questions, answers, and scoring areas, and generate structured assignment data. Specifically, this includes entity recognition through a deep learning model to distinguish different structural blocks, parsing the assignment's logical structure (including the arrangement of multiple-choice options) using a rule engine, generating structured JSON data, and checking the completeness and consistency through a data validation unit to automatically repair missing and conflicting content.

[0035] Step 5: Data Binding and Index Creation Structured job data is bound to dynamic code point data and saved to the database through a data storage unit to establish a job data index. This step includes data association through a mapping unit, secure encoding using an encryption unit, data storage using a distributed storage system, and building a fast query index based on dynamic code points. At the same time, a version control unit records the modification history to ensure data traceability.

[0036] Step Six: Source Tracing Analysis and Report Generation Based on the dynamic code point data query job processing history, modification records and source information, a job traceability report is generated using a traceability analysis algorithm. Specifically, the query unit retrieves the collection time, processing personnel and operation records from the database, uses blockchain technology to write key records into an immutable distributed ledger to enhance credibility, and uses a data analysis unit to statistically analyze the job flow path and efficiency, generating analysis charts and traceability reports.

[0037] Step 7: Visualization and User Interaction The operation traceability report is output in the form of charts, including line graphs, heat maps and time axes, through the visualization unit, and an interactive query interface is provided for users to access. This step includes generating interactive charts using a visualization engine, supporting data filtering by code point, time and personnel through web interface and mobile application, and providing real-time notification of abnormal events through alarm unit, thereby improving user experience and system usability.

[0038] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0039] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for structured data collection and traceability of paper-based operations based on standardized dynamic code points, characterized in that: The method includes the following steps: S1. Generate standardized dynamic code points using a standardized dynamic code point generation device. The standardized dynamic code points contain a unique identifier and are associated with the metadata of the paper work, thereby generating dynamic code point data. S2. Use an image acquisition device to scan the paper work with the standardized dynamic code points attached, obtain work image data, and use a preprocessing unit to perform image enhancement and noise reduction on the work image data to generate preprocessed image data. S3. Based on the dynamic code point data, the identity information of the paper job is parsed, and the job text content is extracted from the preprocessed image data through the optical character recognition unit to generate the original text data; S4. Natural language processing technology is used to perform structured parsing on the original text data, identify the structural elements of the assignment's questions, answers, and scoring areas, and generate structured assignment data. S5. Bind the structured job data with the dynamic code point data, and save it to the database through the data storage unit to establish a job data index; S6. Based on the dynamic code point data, query the processing history, modification records and source information of the job, and generate a job traceability report using the traceability analysis algorithm; S7. Output the job traceability report in chart form through the visualization unit, and provide an interactive query interface for users to access.

2. The paper-based structured data collection and traceability method for work based on standardized dynamic code points according to claim 1, characterized in that, In step S1, standardized dynamic code points are generated using a standardized dynamic code point generation device. Each standardized dynamic code point contains a unique identifier and is associated with metadata from the paper-based work. Generating dynamic code point data includes the following steps: S11. Generate a unique identifier using the standardized dynamic code point generation device and associate it with the metadata of the paper work to generate the dynamic code point data; S12. Using a printing device, the standardized dynamic code dots are attached to a preset position on the paper work, and the size, density, and visibility parameters of the code dots are set through a configuration unit. S13. The dynamic code point data is bound to the metadata of the job creator, creation time, and course information through the metadata association unit to generate code point association data; The dynamic code point generation process is represented by the following formula: ； in, For dynamic code point data, As a unique identifier, For timestamps, As a random factor, This is a generation function based on an encryption algorithm.

3. The method for structured data collection and traceability of paper-based operations based on standardized dynamic code points according to claim 1, characterized in that, In step S2, the image acquisition device scans the paper worksheets with the standardized dynamic code points to obtain worksheet image data. The preprocessing unit then performs image enhancement and noise reduction on the worksheet image data to generate preprocessed image data, including the following steps: S21. Use the image acquisition device to scan the paper work with the standardized dynamic code points attached, obtain the work image data, and adjust the image brightness and contrast through the light compensation unit; S22. The preprocessing unit performs grayscale conversion, binarization, and edge detection on the work image data to remove background noise and distortion. S23. The standardized dynamic code points are located and decoded from the operation image data by the code point recognition unit, the validity of the code points is verified, and the verification result is output to the preprocessed image data.

4. The method for structured data collection and traceability of paper-based operations based on standardized dynamic code points according to claim 1, characterized in that, The steps in S3, which involve parsing the identity information of the paper job based on the dynamic code point data and extracting the job text content from the preprocessed image data using the optical character recognition unit to generate the original text data, include the following: S31. Based on the dynamic code point data, retrieve the associated job metadata from the database and load the job template information; S32. The text region in the preprocessed image data is extracted by the optical character recognition unit, and character segmentation and recognition are performed to generate the original text data. S33. Use the grammar correction unit to perform spell checking and format standardization on the original text data.

5. The method for structured data collection and traceability of paper-based operations based on standardized dynamic code points according to claim 1, characterized in that, The steps in S4, which employ natural language processing technology to perform structured parsing of the original text data, identifying the structural elements of the assignment's questions, answers, and grading areas, and generating structured assignment data, include the following: S41. Use a deep learning model to perform entity recognition on the original text data to distinguish between question, answer, and annotation structure blocks; S42. The logical structure of the job is parsed through the rule engine to generate structured JSON data as the structured job data; S43. Use the data verification unit to check the integrity and consistency of the structured operation data, and automatically repair missing and conflicting content; The structured parsing process is represented by the following formula: ； in, For structured operation data, The original text data, To parse parameters including the recognition thresholds for questions and answers, For deep learning models, This is a parsing function for natural language processing.

6. The method for structured data collection and traceability of paper-based operations based on standardized dynamic code points according to claim 1, characterized in that, In step S5, the structured job data is bound to the dynamic code point data, and saved to the database through the data storage unit. The establishment of the job data index includes the following steps: S51. The structured job data and the dynamic code point data are bound together by a mapping unit, and the data is securely encoded by an encryption unit. S52. Utilize a distributed storage system to save the bound data and establish a fast query index based on dynamic code points; S53. The modification history of the structured job data is recorded through the version control unit to ensure data traceability.

7. The method for structured data collection and traceability of paper-based operations based on standardized dynamic code points according to claim 1, characterized in that, The steps in S6, which involve querying the processing history, modification records, and source information of a job based on the dynamic code point data and generating a job source tracing report using a source tracing analysis algorithm, include the following: S61. Based on the dynamic code point data, query the job processing records in the database, including the collection time, processing personnel, and modification operations; S62. Use blockchain technology to write key operation records into an immutable distributed ledger; S63. The data analysis unit statistically analyzes the workflow and efficiency of the operation and generates analysis charts for the operation traceability report.

8. The method for structured data collection and traceability of paper-based operations based on standardized dynamic code points according to claim 1, characterized in that, The steps in S7, which involve outputting the job traceability report in chart form through a visualization unit and providing an interactive query interface for user access, include the following: S71. The operation traceability report is converted into line chart, heat map and time axis format through the visualization unit; S72. Provide a user query interface using web interfaces and mobile applications, supporting data filtering by code point, time and personnel; S73. Real-time notification of abnormal tracing events is provided through the alarm unit.

9. A paper-based work structured acquisition and traceability system based on standardized dynamic code points, used to implement the paper-based work structured acquisition and traceability method based on standardized dynamic code points as described in any one of claims 1-8, characterized in that, The system includes a dynamic code point generation module, a job acquisition module, a structured processing module, a source analysis module, a visualization feedback module, and an adaptive optimization module; The dynamic code point generation module generates unique dynamic code points using standardized coding units, associates job information through metadata binding units, and outputs code point data through the print control unit. The job acquisition module receives the code point data, acquires the job image through the image acquisition unit, optimizes the image using the preprocessing unit, and verifies the validity of the code points through the code point recognition unit. The structured processing module receives the job image, extracts the text content through the character recognition unit, parses the job structure using the natural language processing unit, and saves the structured data through the data storage unit. The traceability analysis module receives the structured data, retrieves the operation history through the query unit, ensures the data is tamper-proof through the blockchain record unit, and generates a traceability report through the analysis unit. The visualization feedback module receives the source tracing report, creates visualization output through the chart generation unit, provides user interaction through the interface unit, and sends notifications through the alarm unit. The adaptive optimization module receives the output of the source analysis module, evaluates the system efficiency through the performance monitoring unit, optimizes the processing parameters through the parameter adjustment unit, and improves the recognition algorithm through the learning unit.

10. The paper-based work structured collection and traceability system based on standardized dynamic code points according to claim 9, characterized in that, The dynamic code point generation module also includes a code point update unit, which periodically refreshes the dynamic code points to enhance security. The task acquisition module supports collaborative acquisition by multiple devices and enables real-time data upload through a cloud synchronization unit. The structured processing module integrates multiple OCR engines and adaptively selects the optimal algorithm; The source tracing and analysis module supports cross-platform data integration and provides API interfaces for third-party systems to call. The visual feedback module allows for customized report templates and supports multilingual output; The adaptive optimization module dynamically adjusts system parameters based on user feedback.

Images