A real estate electronic certificate management system based on file format adaptation
The real estate electronic certificate management system based on file format adaptation solves the problem of low efficiency in government services caused by inconsistent file formats, and realizes efficient and accurate generation and storage of electronic certificates.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SUZHOU IND PARK SURVEYING MAPPING & GEOINFORMATION CO LTD
- Filing Date
- 2021-10-09
- Publication Date
- 2026-05-19
AI Technical Summary
In existing technologies, differences in data collection equipment, software, and methods lead to inconsistencies in the format of electronic real estate certificates, affecting the efficiency of government services.
The real estate electronic certificate management system adopts a file format adaptive approach, which utilizes a file receiving module, a file format adaptive module, a signing module, and a storage module, combined with a deep neural network to extract and integrate key information, generate electronic certificate files, and affix signatures.
It has improved the accuracy and efficiency of electronic certificate processing, reduced manual data entry errors, and enhanced the efficiency of government services.
Smart Images
Figure CN113947510B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of management system technology, and in particular to a real estate electronic certificate management system based on file format adaptation. Background Technology
[0002] As a government service window with a huge daily workload, real estate registration plays an important role in serving the people and optimizing the business environment. It is also an important indicator for evaluating the city's business environment.
[0003] Currently, government departments need to collect and digitally store supporting documents such as ID cards and household registration books for handling business. However, due to differences in collection equipment, software, and methods, inconsistent file formats can easily occur, making the subsequent processing of electronic certificates more complicated and affecting the efficiency of government services. Summary of the Invention
[0004] Therefore, the technical problem to be solved by the present invention is to overcome the problem that the different data acquisition devices, data acquisition software and data acquisition methods in the prior art lead to inconsistent file formats, which reduces the efficiency of government affairs, and to provide a real estate electronic certificate management system with high processing efficiency based on file format adaptation.
[0005] To address the aforementioned technical problems, this invention provides a real estate electronic certificate management system based on file format adaptation, comprising:
[0006] A file receiving module, which is used to receive files;
[0007] The file receiving module transmits the received file to the file format adaptation module, and the file format adaptation module generates a new electronic certificate file.
[0008] The signature module, in which the file format adaptive module transmits the generated new electronic certificate file to the signature module, is used to generate the corresponding electronic signature file;
[0009] The storage module is used to receive electronic certificate documents and electronic signature documents from the signature module and store relevant information.
[0010] As a further improvement of the present invention, the file receiving module receives files via a REST interface.
[0011] As a further improvement of the present invention, the file format adaptive module uses Apache POI to read the content of Microsoft Office Word format files and extracts the content of the files by using an end-to-end neural network model. Apache POI is a Java library for parsing Microsoft Documents.
[0012] As a further improvement of the present invention, the file format adaptive module reads the content of PDF files using the itextpdf method and extracts the content of the files using an end-to-end neural network model. Here, itextpdf is a Java library for parsing PDF files.
[0013] As a further improvement of the present invention, the end-to-end neural network model uses ResNet+FPN as the backbone network for text detection, and uses the results of the detection network and ROIAlign algorithm to perform attention+encoder-decoder text recognition, fusing text position information, visual information and semantic information, and using BiLSTM for IOB label classification to obtain key document information, where attention is the attention mechanism, encoder is the encoder, and decoder is the decoder.
[0014] As a further improvement of this invention, the end-to-end neural network model performs joint optimization learning on the three tasks of text detection, text recognition, and information extraction, and its loss function is:
[0015]
[0016] Where, λ recog and λ info For balance coefficient, The loss function for the text detection task includes a classification loss function and a regression loss function.
[0017] As a further improvement of the present invention, the text recognition loss function and the information extraction loss function are defined as follows:
[0018]
[0019]
[0020] in, Let be the true label of the t-th character in the i-th text from the recognition task. These are the corresponding information extraction tags.
[0021] As a further improvement of the present invention, text detection and text recognition are combined into a text reading task, and multimodal visual and text features of text reading are fused into information extraction through a unified end-to-end text reading and information extraction network.
[0022] As a further improvement of the present invention, the signing module uses the SHA-256 algorithm to calculate the HASH value of the electronic certificate file, and uses the RSA algorithm to sign the calculated HASH value with a private key, and finally generates the corresponding electronic signature file.
[0023] As a further improvement of the present invention, the relevant information is a set of information formed by the storage module after receiving the electronic certificate documents and electronic signature documents from the signature module, extracting features from the certificates according to the data specifications of real estate ownership, real estate unit, and certificate serial number, and sorting and encoding the information, and storing the set of information in the database.
[0024] The technical solution of the present invention has the following advantages compared with the prior art:
[0025] The present invention discloses a real estate electronic certificate management system based on file format adaptation. This system extracts key information from the content through a deep neural network and integrates it in a unified manner, thereby improving accuracy and efficiency. It extracts key information of real estate electronic certificates, such as real estate unit and certificate serial number, from the content of files in different formats. This eliminates the need for manual reading and data entry, improving work efficiency and reducing subsequent problems caused by manual entry errors. Attached Figure Description
[0026] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein:
[0027] Figure 1 This is an architecture diagram of a real estate electronic certificate management system based on file format adaptation in a preferred embodiment of the present invention. Detailed Implementation
[0028] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.
[0029] In some embodiments, refer to Figure 1 As shown, the present invention provides a real estate electronic certificate management system based on file format adaptation, comprising:
[0030] The file receiving module is used to receive files.
[0031] The file format adaptation module receives the file and passes it to the file format adaptation module, which then generates a new electronic certificate file.
[0032] The signature module and the file format adaptation module pass the newly generated electronic certificate file to the signature module, which then generates the corresponding electronic signature file.
[0033] The storage module receives electronic certificate documents and electronic signature documents from the signature module and stores relevant information.
[0034] After receiving a file, the file receiving module passes it to the file format adaptation module. The file format adaptation module generates a new electronic certificate file and passes it to the signing module. The signing module generates a corresponding electronic signature file. The storage module receives the electronic certificate file and electronic signature file from the signing module and stores the relevant information.
[0035] In some embodiments, refer to Figure 1 As shown, the file receiving module receives files via a REST interface. REST is used to standardize how applications interact with API providers at the HTTP layer; REST describes the rules for data interaction between clients and servers at the HTTP layer; the client completes an HTTP interaction by sending an HTTP request to the server and receiving the server's response. In this interaction, the REST architecture defines two important aspects: the method used for the HTTP request and the request link.
[0036] In some embodiments, refer to Figure 1 As shown, the file format adaptation module uses Apache POI to read the content of Microsoft Office Word format files. It extracts the content using an end-to-end neural network model. Apache POI is a Java library for parsing Microsoft Documents. It extracts the property unit and certificate serial number information, then reorganizes the content according to the electronic certificate template to generate a new electronic certificate document.
[0037] In some embodiments, refer to Figure 1 As shown, the file format adaptation module uses the `itextpdf` method to read the content of PDF files and extracts the content using an end-to-end neural network model. `itextpdf` is a Java library for parsing PDF files. It extracts the real estate unit and certificate serial number information, and then reorganizes and formats the content according to the electronic certificate template to generate a new electronic certificate file.
[0038] In some embodiments, refer to Figure 1 As shown, the end-to-end neural network model uses ResNet+FPN as the backbone network for text detection, and uses the results of the detection network and ROIAlign algorithm for attention+encoder-decoder text recognition. It integrates text location information, visual information, and semantic information, and uses BiLSTM for IOB label classification to obtain key document information. Here, attention is the attention mechanism, encoder is the encoder, and decoder is the decoder.
[0039] In some embodiments, refer to Figure 1 As shown, the end-to-end neural network model performs joint optimization learning on three tasks: text detection, text recognition, and information extraction. Its loss function is:
[0040]
[0041] Where, λ recog and λ info For balance coefficient, The loss function for the text detection task includes a classification loss function and a regression loss function.
[0042] In some embodiments, refer to Figure 1 As shown, the text recognition loss function and the information extraction loss function are defined as follows:
[0043]
[0044]
[0045] in, Let be the true label of the t-th character in the i-th text from the recognition task. These are the corresponding information extraction tags.
[0046] It should be noted that the algorithm achieves an accuracy of 91% and 88% on the Taxi Invoices and Resumes datasets, respectively, which is an improvement in accuracy compared to similar algorithms. Specific dataset test data are shown in Tables 1 and 2.
[0047]
[0048]
[0049] Table 1. Accuracy of information extraction from the TaxiInvoices dataset (%)
[0050] Extracting Projects / Algorithms Chargrid NER GCN This application Name 43.4 42.7 42.9 45.7 Phone number 87.0 86.6 83.3 88.0 E-mail 70.9 69.6 68.0 74.9 age 77.1 68.7 62.2 81.4 School 74.7 86.0 82.3 87.4
[0051] Table 2. Accuracy of information extraction from the Resumes dataset (%)
[0052] In some embodiments, refer to Figure 1 As shown, text detection and text recognition are combined into a text reading task. A unified end-to-end text reading and information extraction network integrates multimodal visual and textual features from text reading into information extraction. While the multimodal visual and textual features from text reading are fused into information extraction, the semantics in information extraction also contribute to the optimization of text reading, thereby improving the accuracy of document information extraction.
[0053] In some embodiments, refer to Figure 1 As shown, the signing module uses the SHA-256 algorithm to calculate the HASH value of the electronic certificate file, and uses the RSA algorithm to sign the calculated HASH value with a private key, ultimately generating the corresponding electronic signature file.
[0054] In some embodiments, refer to Figure 1 As shown, the relevant information is a set of information formed by the storage module after receiving the electronic certificate and electronic signature documents from the signature module, extracting features from the certificates according to the data specifications of real estate ownership, real estate unit, and certificate serial number, and sorting and encoding the information, and storing the set of information in the database.
[0055] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0056] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0057] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0059] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.
Claims
1. A real estate electronic certificate management system based on file format adaptation, characterized in that, include: A file receiving module, which is used to receive files; The file receiving module transmits the received file to the file format adaptation module, and the file format adaptation module generates a new electronic certificate file. The signature module, in which the file format adaptive module transmits the generated new electronic certificate file to the signature module, is used to generate the corresponding electronic signature file; A storage module is provided for receiving electronic certificate documents and electronic signature documents from the signature module and storing relevant information. For PDF files, the file format adaptive module uses the iftextpdf method to read the content of the file and extracts the content of the file by using an end-to-end neural network model. iftextpdf is a Java library for parsing PDF files. The end-to-end neural network model uses ResNet+FPN as the backbone network for text detection, and uses the results of the detection network and ROIAlign algorithm to perform text recognition using attention+encoder-decoder. It integrates text location information, visual information, and semantic information, and uses BiLSTM for IOB label classification to obtain key document information. Here, attention is the attention mechanism, encoder is the encoder, and decoder is the decoder.
2. The real estate electronic certificate management system based on file format adaptation according to claim 1, characterized in that, The file receiving module receives files via a REST interface.
3. The real estate electronic certificate management system based on file format adaptation according to claim 1, characterized in that, For Microsoft Office Word format files, the file format adaptive module uses Apache POI to read the file content and extracts the content using an end-to-end neural network model. Apache POI is a Java library for parsing Microsoft Documents.
4. The real estate electronic certificate management system based on file format adaptation according to claim 1, characterized in that, The end-to-end neural network model performs joint optimization learning on three tasks: text detection, text recognition, and information extraction. Its loss function is: ； in, and For balance coefficient, The loss function for the text detection task includes a classification loss function and a regression loss function.
5. A real estate electronic certificate management system based on file format adaptation according to claim 4, characterized in that, The text recognition loss function and the information extraction loss function are defined as follows: ； ； in, Let be the true label of the t-th character in the i-th text from the recognition task. These are the corresponding information extraction tags.
6. The real estate electronic certificate management system based on file format adaptation according to claim 4, characterized in that, Text detection and text recognition are combined into a text reading task, which integrates multimodal visual and text features of text reading into information extraction through a unified end-to-end text reading and information extraction network.
7. The real estate electronic certificate management system based on file format adaptation according to claim 1, characterized in that, The signing module uses the SHA-256 algorithm to calculate the HASH value of the electronic certificate file, and uses the RSA algorithm to sign the calculated HASH value with a private key, ultimately generating the corresponding electronic signature file.
8. The real estate electronic certificate management system based on file format adaptation according to claim 1, characterized in that, The relevant information is a set of information formed by the storage module after receiving the electronic certificate and electronic signature documents from the signature module, extracting features from the certificates according to the data specifications of real estate ownership, real estate unit, and certificate serial number, and sorting and encoding the information, and storing the set of information in the database.