Qualification verification data management system

Abstract

The invention discloses a qualification verification data management system, and relates to the technical field of operation and maintenance management, and the system comprises a mobile terminal which is used for receiving to-be-verified qualification information. And the verification attribute extraction module is used for connecting the open database to obtain the certificate template according to the target qualification type, and identifying character attributes and format layout characteristics. And the check tree construction module is used for constructing a check binary tree. The RPA robot comprises a verification work order configuration module and is used for receiving the to-be-verified qualification information from the multiple mobile terminals, receiving the multiple verification binary trees from the verification tree construction module and filling the verification binary trees with the to-be-verified qualification information to generate to-be-verified work order binary trees; and the qualification verification module is used for accessing the multi-open database according to the target qualification type and carrying out information comparison on the plurality of work order binary trees to be verified so as to achieve the technical effects of improving the verification efficiency, enhancing the verification accuracy and optimizing the adaptive capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance management, and particularly to a qualification verification data management system. Background Art

[0002] Qualification verification is an important link to ensure that cooperative units and relevant personnel meet the requirements. Traditional qualification verification methods mainly rely on manual verification, and staff need to check the validity, authenticity, and integrity of qualification certificates one by one. However, with the expansion of enterprise business and the increase in the scale of cooperation, the workload of qualification verification has been continuously increasing, and the manual verification method has gradually revealed defects such as low efficiency and easy errors. Summary of the Invention

[0003] In view of the technical problems of low verification efficiency, low accuracy, and poor adaptability in the prior art, the present invention provides a qualification verification data management system to solve these problems.

[0004] The technical solution of the present invention to solve the above technical problems is as follows: The present invention provides a qualification verification data management system, including: A mobile terminal, which is used to receive the qualification information to be verified of the user.

[0005] A verification attribute extraction module, which is used to access an open database according to the target qualification type, obtain a certificate template, and identify a set of text attributes and a set of format layout features.

[0006] A verification tree construction module, which is used to construct a verification binary tree according to the set of text attributes and the set of format layout features.

[0007] An RPA robot, including: A verification work order configuration module, which is used to receive multiple qualification information to be verified from multiple mobile terminals, receive multiple verification binary trees from the verification tree construction module, and fill the multiple qualification information to be verified into the multiple verification binary trees to generate multiple verification work order binary trees.

[0008] A qualification verification module, which is used to access multiple open databases according to multiple target qualification types, respectively compare the information of the multiple verification work order binary trees, and when they match, send qualification verification completion information to the corresponding mobile terminal, otherwise, send qualification verification failure information to the corresponding mobile terminal.

[0009] The beneficial effects of the present invention are as follows: receiving the qualification information to be verified submitted by the user through the mobile terminal; the verification attribute extraction module accesses the open database according to the target qualification type, obtains the certificate template, and identifies the set of text attributes and the set of format layout features therein; the verification tree construction module constructs a verification binary tree according to the extracted set of text attributes and the set of format layout features; the RPA robot includes: a verification work order configuration module, which receives the qualification information to be verified from multiple mobile terminals, receives the verification binary tree from the verification tree construction module, fills the qualification information to be verified into the corresponding verification binary tree, and generates multiple verification work order binary trees; the qualification verification module accesses multiple open databases according to the target qualification type, compares the information of the verification work order binary trees respectively, if they match, it sends the qualification verification completion information to the corresponding mobile terminal, otherwise it sends the qualification verification failure information, thereby achieving the technical effects of improving the verification efficiency, enhancing the verification accuracy, and optimizing the adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a schematic structural diagram of a qualification verification data management system of the present invention; Figure 2 It is an exemplary visualization result diagram of the visualization end in a qualification verification data management system of the present invention.

[0011] In the drawings, the components represented by each reference numeral are described as follows: Mobile terminal 11, verification attribute extraction module 12, verification tree construction module 13, RPA robot 14, verification work order configuration module 141, qualification verification module 142. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0013] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more, unless otherwise specifically defined.

[0014] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to implement and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without the use of these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0015] Figure 1 FIG. 0 is a schematic structural diagram of a qualification verification data management system of the present invention, which includes: A mobile terminal 11, configured to receive qualification information to be verified of a user.

[0016] Specifically, the mobile terminal 11 refers to a mobile device used by a user, such as a smart phone, a tablet computer, etc., and the mobile terminal 11 interacts with the intelligent operation and maintenance system through a mobile application or a mobile browser.

[0017] Specifically, the qualification information to be verified refers to files or data uploaded by the mobile terminal through a user interface for proving that a user or a unit has a certain qualification or meets a certain standard, including various forms such as text, pictures, and videos. Exemplarily, it includes certificates, licenses, etc.

[0018] In some embodiments, when receiving the qualification information to be verified of the user, the execution steps of the mobile terminal 11 include: Obtaining initial upload information, where the initial upload information includes at least one of text, pictures, and videos; when the initial upload information is a picture, identifying picture text information and picture format layout information; when the initial upload information is a video, extracting video text information and video frame picture format layout information; taking the union of the text, the picture text information, the picture format layout information, the video text information, and the video frame picture format layout information to obtain union text information and union picture format layout information, and storing them as the qualification information to be verified.

[0019] Specifically, the mobile terminal obtains the initial upload information, which is the original information entered by the user and includes at least one of text, pictures, or videos.

[0020] Specifically, when the initial uploaded information is a picture, optical character recognition (OCR) technology is used to recognize the text information in the picture, and at the same time, the format layout of the picture is analyzed, including the position, size, direction of the text, and the overall typesetting of the picture, etc., so as to convert the text in the image or paper document into an editable and searchable digital format to ensure accurate comparison during subsequent verification. For example, for a picture of a business license, the system will recognize the text information such as the enterprise name, registration number, legal representative, etc., and record the position and typesetting method of these texts in the picture.

[0021] Specifically, when the initial uploaded information is a video, first, the video is decomposed into frames, the video is split into a series of static picture frames, and pairwise similarity evaluation is performed on the video frame pictures to remove duplicate or highly similar frames to reduce the data volume and improve the processing efficiency; then, the text information and format layout information of the remaining video frame pictures are recognized to form complete video text information and video frame picture format layout information. Among them, for a video without subtitles, speech recognition technology can be used to convert the speech information in the video into text information.

[0022] Furthermore, the content extracted from the text, picture text information, picture format layout information, video text information, and video frame picture format layout information is merged, taking the union, to obtain union text information and union picture format layout information, which are stored as the qualification information to be verified for subsequent processing and verification.

[0023] In some implementation manners, when the initial uploaded information is a video, to extract the video text information and video frame picture format layout information, the execution steps of the mobile device 11 further include: Decompose the video into frames to obtain a set of video frame pictures; perform pairwise similarity evaluation on the set of video frame pictures to obtain multiple video frame picture similarities; based on the multiple video frame picture similarities, store the pictures with a similarity equal to 1 one by one to obtain an updated set of video frame pictures; recognize the set of picture text information and the set of picture format layout information of the updated set of video frame pictures, and store them as video text information and video frame picture format layout information.

[0024] Exemplarily, first, use a video processing library such as FFmpeg to decompose the video into a series of consecutive frame images, and output them as a set of video frame pictures. Each frame in this set of video frame pictures represents an instantaneous picture in the video; then, calculate the similarity between each pair of video frames to identify duplicate or similar frames. Among them, the similarity evaluation methods include: evaluating image similarity by generating the hash value of the image, measuring the similarity of two images in terms of brightness, contrast, and structure through the structural similarity index, and calculating the sum of the squares of the differences in pixel values of the two images. The smaller the difference, the higher the similarity.

[0025] Exemplarily, for frames with a similarity of 1 (i.e., completely identical frames), one of them is retained, and the remaining duplicate frames are removed to obtain an updated set of video frame pictures, thereby reducing redundant data and improving the efficiency of subsequent processing. Further, each frame picture in the updated set of video frame pictures is identified for text and format layout information through technologies such as optical character recognition (OCR), converting the text in the image into editable and searchable text, identifying structural information such as titles, paragraphs, and tables, and forming a set of video text information and video frame picture format layout information, thereby providing a comprehensive visual information basis for subsequent qualification verification.

[0026] The above steps effectively remove duplicate picture frames through similarity evaluation and one-retained storage of video frame pictures, reduce the number of pictures to be processed, save storage space and computing resources, and improve the efficiency of subsequent processing. Each frame in the updated set of video frame pictures has unique information or perspectives, thus helping to more comprehensively reflect the appearance and content changes of the qualification certificate.

[0027] In some implementation manners, when the initial uploaded information is a picture, identifying the picture text information and the picture format layout information, the execution steps of the mobile device 11 further include: Processing the picture through a text information extraction model to obtain the picture text information, wherein the text information extraction model is generated by training a convolutional neural network with multiple groups of data, and any one of the multiple groups of data includes picture data and text labels obtained by processing the picture data through OCR; processing the picture through a picture table layout extraction model to obtain the picture table layout information, wherein the picture table layout extraction model is generated by training a convolutional neural network with multiple groups of data, and any one of the multiple groups of data includes picture data and labels annotating the table structure features; processing the picture through a picture text layout extraction model to obtain the picture text layout information, wherein the picture text layout extraction model is generated by training a convolutional neural network with multiple groups of data, and any one of the multiple groups of data includes picture data and labels annotating the text layout positions.

[0028] Specifically, the text information extraction model is constructed based on a convolutional neural network and is used to extract text information from pictures. This model learns through a large amount of training data, where the training data includes picture data (original pictures) and their corresponding text labels. For example, a picture in the training data may be a business license, and its corresponding text label is the enterprise name, registration number, and other text information on the business license.

[0029] Specifically, a picture table layout extraction model is also constructed and trained based on a convolutional neural network to identify and extract table structure features from pictures. Among them, the training data includes picture data (original pictures) and table structure feature labels (annotating the features of the table structure in the pictures, such as rows, columns, cells, etc.). For example, for a picture containing a financial statement, its label will annotate the table frame features of the statement, the titles of each row and column, and other contents.

[0030] Specifically, similarly, the picture text layout extraction model is also trained using a convolutional neural network to specifically identify the layout positions of the text in the pictures. The training data of the picture text layout extraction model consists of pictures and labels annotating the layout positions of the text. Among them, the labels clarify the specific positions of each text or text block in the pictures, such as the coordinates, sizes, and other information of the text. For example, for a picture containing multiple paragraphs of text descriptions, the label will annotate the specific position coordinates of each paragraph of text in the picture (such as the diagonal coordinates of the recognition box).

[0031] In the above steps, through a specifically trained convolutional neural network model, the text content, table structure, and text layout in the pictures can be identified more accurately, reducing recognition errors and omissions compared to OCR technology, improving the accuracy and reliability of information extraction, while reducing manual intervention and improving the automation level and processing efficiency of the system.

[0032] The verification attribute extraction module 12 is used to access an open database according to the target qualification type, obtain a certificate template, and identify the set of text attributes and the set of format layout features.

[0033] Specifically, the verification attribute extraction module 12 accesses the corresponding open database or public resource library according to the target qualification type by using API interfaces or database connection technologies, such as RESTful API, SOAP, etc., to obtain a standardized certificate template. Among them, the target qualification type refers to the type of qualification certificate to be verified, such as business license, ID card, driver's license, etc. Different qualification types have different certificate templates and attributes. The certificate template refers to the pre-defined certificate format, including fixed text content and layout.

[0034] Specifically, the set of text attributes is a set of attributes of all text information extracted from the certificate template, including attribute labels such as certificate number, name, expiration date, etc. The set of format layout features is a set of format and layout information extracted from the certificate template, including the position, size, font, color of the text, and the layout information of elements such as tables and graphics. Among them, the set of text attributes and the set of format layout features have an associated relationship, that is, through the set of text attributes and the set of format layout features, the text attributes in the certificate template and their corresponding position ranges can be determined for comparison with the qualification certificate to be verified in the subsequent verification process.

[0035] By obtaining the certificate template from the open database, the verification attribute extraction module can accurately identify the key attributes and format layout features in the qualification certificate, reducing verification errors caused by template mismatch or inaccurate attribute extraction. At the same time, it can dynamically access the corresponding certificate template according to different target qualification types, adapt to the verification requirements of various types of qualification certificates, and enhance the versatility and adaptability of the system.

[0036] The verification tree construction module 13 is used to construct a verification binary tree according to the set of text attributes and the set of format layout features.

[0037] Specifically, the function of the verification tree construction module is to construct a verification binary tree according to the set of text attributes and the set of format layout features, so that when an abnormal node appears, an error can be immediately returned, thereby greatly improving the efficiency of the system.

[0038] Specifically, the binary tree is constructed recursively starting from the root node. Each node in the structure of the binary tree represents a verification point, including the corresponding attributes and layout information. The nodes are connected by pointers or references until all verification points are included in the tree. During the verification process, each node of the binary tree is traversed to check whether its corresponding attributes and layout meet the expectations. Once it is found that the verification of a certain node fails, the traversal is immediately terminated and an error message is returned.

[0039] In some embodiments, a verification binary tree is constructed according to the set of text attributes and the set of format layout features. The execution steps of the verification tree construction module 13 include: The set of format layout features includes a set of table structure features and a set of text attribute layout positions. A first verification binary subtree is constructed according to the set of text attributes; a second verification binary subtree is constructed according to the set of table structure features; a third verification binary subtree is constructed according to the set of text attribute layout positions; the first verification binary subtree, the second verification binary subtree, and the third verification binary subtree are added to the verification binary tree.

[0040] Specifically, the set of table structure features refers to the set of table-related feature information extracted from the certificate template. Exemplarily, it includes the number of rows and columns of the table, the cell content, the border style of the table, etc. The set of text attribute layout positions refers to the set of specific position information of the text in the certificate template. Exemplarily, it includes the coordinate position of the text, the distance between the texts, the position of the text relative to the table or other elements, etc.

[0041] Specifically, a first verification binary subtree is constructed according to the set of text attributes. Each node in this binary tree represents a verification point of a text field, and the nodes are connected by pointers or references to verify the text content in the qualification certificate.

[0042] Specifically, according to the same method principle of constructing the first verification binary tree, the second verification binary tree and the third verification binary tree are respectively constructed based on the table structure feature set and the text attribute layout position set. Preferably, the verification point order of the third verification binary tree is the same as that of the first verification binary tree, that is, the first verification binary tree and the third verification binary tree perform verification operations in the same order. For example, for a business license, the unified social credit code is used as the root node for verification, and text attributes such as name, type, legal representative, etc. are used as child nodes.

[0043] Furthermore, the above three verification binary trees are integrated into a complete verification binary tree to form a multi-level and multi-dimensional verification structure, so as to comprehensively verify the text content, table structure and text layout position of the qualification certificate.

[0044] The above process constructs independent verification subtrees for text attributes, table structure features and text layout positions respectively, and integrates them into the verification binary tree, which can comprehensively and meticulously verify the qualification certificate, reduce verification errors caused by incomplete or inaccurate features in a certain aspect, improve the accuracy of verification. Once an abnormality is found at a certain node, the error information can be immediately returned, and the subsequent verification can be stopped, saving verification time and improving verification efficiency.

[0045] In some implementation manners, according to the text attribute set, the first verification binary tree is constructed. The execution steps of the verification tree construction module 13 further include: Extracting the first text attribute of the text attribute set and setting it as the root node of the binary tree; extracting the second text attribute of the text attribute set and setting it as the second-level node of the binary tree; until the Nth text attribute of the text attribute set is extracted and set as the Nth-level node of the binary tree; connecting the root node of the binary tree, the second-level node of the binary tree until the Nth-level node of the binary tree to generate the first verification binary tree.

[0046] Specifically, the verification tree construction module starts from the first text attribute in the text attribute set, followed by the second text attribute, until the Nth text, and sets them as the root node, second-level node, third-level node of the binary tree, until the Nth-level node, to generate the hierarchical first verification binary tree.

[0047] Exemplarily, for a business license, first, the unified social credit code is used as the first text attribute for verification and the root node is set; then, text attributes such as name, type, legal representative, etc. are used as the second-level node, third-level node, until the Nth-level node respectively to form the corresponding first verification binary tree.

[0048] By constructing the first verification binary tree, the hierarchical order for verifying the text attributes of qualification certificates is defined, which helps to detect subtle errors and inconsistencies in qualification certificates, thereby improving the accuracy of verification.

[0049] The RPA robot 14 includes: A verification work order configuration module 141, which is used to receive multiple qualification information to be verified from multiple mobile terminals 11, receive multiple verification binary trees from the verification tree construction module 13, and fill the multiple qualification information to be verified into the multiple verification binary trees to generate multiple work order binary trees to be verified.

[0050] Specifically, RPA, that is, Robotic Process Automation, is used to simulate human beings for automated operations. In this system, the RPA robot 14 is used to handle the entire process of qualification verification.

[0051] Specifically, the RPA robot 14 includes a verification work order configuration module 141 responsible for receiving the qualification information to be verified from the mobile terminal and the verification binary tree from the verification tree construction module, and integrating the two to generate a work order binary tree to be verified, and a qualification verification module 142 that accesses the open database according to the target qualification type, and compares the information of multiple work order binary trees to be verified, and finally sends the information of verification completion or failure to the mobile terminal according to the comparison result.

[0052] Specifically, the verification work order configuration module 141 has multi-thread processing capabilities. First, it receives multiple qualification information to be verified from multiple mobile terminals 11, and these information may come from different users or enterprises; then, it analyzes the qualification types of the multiple qualification information to be verified, and obtains multiple verification binary trees from the verification tree construction module 13 based on the analysis results, and each verification binary tree corresponds to a specific qualification type; then, it fills the received qualification information to be verified into the corresponding verification binary tree nodes to generate multiple work order binary trees to be verified, and each work order corresponds to a specific qualification to be verified.

[0053] Exemplarily, if a qualification information to be verified is a business license, the verification work order configuration module 141 will fill this information into the verification binary tree corresponding to the business license, including matching each attribute and feature in the qualification information with the nodes of the verification binary tree, and storing the information in the corresponding nodes.

[0054] Through the verification work order configuration module 141, the scattered qualification information to be verified and the verification rules can be integrated to form an orderly verification work order, which helps the subsequent verification process to proceed according to the established rules and order, improving the efficiency and accuracy of verification. For example, in an engineering project, it may be necessary to verify the qualifications of multiple suppliers. The verification work order configuration module 141 can quickly organize the qualification information of multiple suppliers into a binary tree of work orders to be verified, facilitating subsequent unified verification.

[0055] The qualification verification module 142 is used to access multiple open databases according to multiple target qualification types, and respectively compare the information of the multiple binary trees of work orders to be verified. When they match, it sends a qualification verification completion message to the corresponding mobile terminal 11; otherwise, it sends a qualification verification failure message to the corresponding mobile terminal 11.

[0056] Specifically, the open databases contain data such as standard information, templates of various qualification certificates, and relevant verification rules. The qualification verification module 142 needs to access the corresponding open database according to the target qualification type to obtain the reference information required for verification.

[0057] Specifically, the qualification verification module 142 first accesses multiple open databases according to multiple target qualification types. For example, when verifying multiple enterprise business licenses, the module will access the open database containing the standard information and templates related to business licenses. Then, it respectively compares the multiple binary trees of work orders to be verified with the information in the open database, including checking whether the text content in the text attribute set is consistent with the standard text content in the database, and whether the table structure and text layout position in the format layout feature set conform to the standard format layout in the database. If all comparisons match, it means the qualification verification is passed. At this time, the qualification verification module 142 sends a qualification verification completion message to the corresponding mobile terminal 11. On the contrary, if there is any mismatch in any link of the comparison, it means the qualification verification fails, and a qualification verification failure message is sent to the mobile terminal 11 accordingly.

[0058] By docking with the open database, the qualification verification module 142 can timely update the verification standards and rules, ensuring that the verification results are consistent with the latest qualification requirements and qualification holding status, ensuring the efficiency and accuracy of verification, and improving the standardization level of qualification certification.

[0059] In some embodiments, when accessing multiple open databases according to multiple target qualification types and respectively comparing the information of the multiple binary trees of work orders to be verified, the execution steps of the qualification verification module 142 include: Access the first open database according to the first target qualification type, and extract the first work order binary tree to be verified; obtain the root node data of the first work order binary tree to be verified; when the root node data is a layout feature, compare the qualification information to be verified stored in the root node with the format layout feature. If they are consistent, the root node data is considered to match. If they are inconsistent, send a qualification verification failure message to the corresponding mobile device 11; when the root node data is text, search for the text in the first open database. If it is retrieved, the root node data is considered to match. If it is not retrieved, send a qualification verification failure message to the corresponding mobile device 11; when the root node data matches, obtain the secondary node data of the first work order binary tree to be verified and continue the verification; until each node of the first work order binary tree to be verified is verified to match, it is considered to match.

[0060] Specifically, the first target qualification type can be any one of multiple target qualification types or the first one based on a preset sorting rule (such as submission time sequence, qualification importance, importance of the entity corresponding to the mobile device 11, etc.); the first open database is a database containing standard information, templates, and verification rules related to the first target qualification type.

[0061] Specifically, first, access the first open database according to the first target qualification type, then, extract the first work order binary tree to be verified from the work order verification configuration module 141; next, verify the root node data of the first work order binary tree to be verified.

[0062] Among them, if the root node data is a layout feature, compare the format layout feature of the qualification information to be verified stored in the root node with the standard format layout feature in the first open database; for example, when verifying a business license, the root node may be layout features such as the title position and font size of the business license. If these features are consistent with the standards in the database, the root node data matches; if they are inconsistent, it is considered that the verification fails, and a qualification verification failure message is sent to the corresponding mobile device 11.

[0063] Among them, if the root node data is text, search for the text in the first open database; for example, in a business license, the root node may be the specific "Unified Social Credit Code". If this text is retrieved in the database, it indicates that the root node data matches; if it is not retrieved, it means that the code does not exist, and it is considered that the verification fails, and a failure message is sent to the mobile device 11.

[0064] Further, if the root node data matches, continue to obtain the secondary node data of the first work order binary tree to be verified for verification, and progress layer by layer until all nodes are verified; for example, the secondary node of the business license may be the enterprise name, and the tertiary node is the registration number, etc. Only when each node in the binary tree is verified to match, the qualification verification is regarded as a whole match. At this time, the qualification verification module 142 sends a qualification verification completion message to the mobile terminal. Correspondingly, if any node is found to not match, the entire qualification verification will be judged as a failure, and the verification will stop and a failure message will be sent to the mobile terminal 11.

[0065] In some embodiments, as Figure 2 shown, the RPA robot 14 further includes: A visualization terminal, configured to display the user and work order verification status on a visualization interface when multiple work orders binary trees to be verified are constructed. The work order verification status includes not started, in progress, and completed.

[0066] Specifically, the visualization terminal is a functional component in the RPA robot 14 that provides a user-friendly graphical interface to display the running status and relevant information of the system in real time, ensuring that users can intuitively understand the verification progress and results of the work orders.

[0067] Specifically, when multiple work orders binary trees to be verified are constructed, the visualization terminal will display the verification status of each work order on its interface; for example, for work order 1, its verification has not been started, and the visualization terminal shows "not started"; when the verification is in progress, the status will be updated to "in progress"; once the verification is completed, the status will be shown as "completed".

[0068] Through the real-time status display, users can intuitively understand the processing progress of each work order, timely discover which work orders have not been processed, which are in progress, and which have been completed, thereby helping to improve work efficiency and ensuring that all work orders can be processed in a timely manner.

[0069] In summary, the qualification verification data management system provided by the present invention has the following technical effects: Receive the qualification information to be verified submitted by the user through the mobile terminal; the verification attribute extraction module accesses the open database according to the target qualification type, obtains the certificate template, and identifies the set of text attributes and the set of format layout features therein; the verification tree construction module constructs a verification binary tree according to the extracted set of text attributes and the set of format layout features; the RPA robot includes: a verification work order configuration module, which receives the qualification information to be verified from multiple mobile terminals, receives the verification binary tree from the verification tree construction module, fills the qualification information to be verified into the corresponding verification binary tree, and generates multiple verification work order binary trees; the qualification verification module accesses multiple open databases according to the target qualification type, and performs information comparison on the verification work order binary trees respectively. If they match, it sends a qualification verification completion message to the corresponding mobile terminal, otherwise it sends a qualification verification failure message, thereby achieving the technical effects of improving the verification efficiency, enhancing the verification accuracy, and optimizing the adaptability.

[0070] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0071] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented 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.

[0072] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0073] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions in the process Figure 1one or more processes and / or blocks Figure 1 functions specified in one or more blocks.

[0074] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, so that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable apparatus provide steps for realizing the functions specified in one Figure 1 one or more processes and / or blocks Figure 1 or more processes and / or blocks.

[0075] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic inventive concept.

[0076] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A qualification verification data management system, characterized in that: include: The mobile terminal is used to receive the user's qualification information to be verified; The verification attribute extraction module is used to access the open database, obtain the certificate template, and identify the text attribute set and format layout feature set according to the target qualification type; A verification tree construction module is used to construct a verification binary tree according to a text attribute set and a format layout feature set; RPA robots include: A verification work order configuration module is used to receive multiple qualification information to be verified from multiple mobile terminals, receive multiple verification binary trees from the verification tree construction module, and fill the multiple qualification information to be verified into the multiple verification binary trees to generate multiple binary trees of work orders to be verified; The qualification verification module is used to access multiple open databases according to multiple target qualification types, and compare the information of the multiple binary trees of work orders to be verified respectively. When they match, the qualification verification completion information is sent to the corresponding mobile terminal, otherwise, the qualification verification failure information is sent to the corresponding mobile terminal.

2. The system according to claim 1, characterized in that Receiving the user's qualification information to be verified, the mobile terminal executes the steps including: Obtaining initial upload information, where the initial upload information includes at least one of text, image, and video; When the initial uploaded information is a picture, identifying the picture text information and picture format layout information; When the initial uploaded information is a video, extract the video text information and the video frame picture format layout information; Take the union of text, picture text information, picture format layout information, video text information and video frame picture format layout information to obtain the union text information and the union picture format layout information, and store them as the qualification information to be verified.

3. The system according to claim 2, characterized in that When the initial upload information is a video, extracting video text information and video frame picture format layout information, the execution step of the mobile terminal further includes: Decompose the video into frames to obtain a set of video frame images; Perform pairwise similarity evaluation on the video frame image set to obtain the similarity of multiple video frame images; Based on the similarities of multiple video frame images, pictures with a similarity equal to 1 are left-one-out stored to obtain a video frame image update set; A picture text information set and a picture format layout information set of a video frame picture update set are identified and stored as video text information and video frame picture format layout information.

4. The system according to claim 2, characterized in that When the initial uploaded information is a picture, identifying the picture text information and the picture format layout information, the execution step of the mobile terminal further includes: Processing the image through a text information extraction model to obtain text information of the image, wherein the text information extraction model is generated by training a convolutional neural network through multiple sets of data, and any one of the multiple sets of data includes the image data and the text label obtained by processing the image data through OCR; Processing the image through a picture table layout extraction model to obtain picture table layout information, wherein the picture table layout extraction model is generated by training a convolutional neural network with multiple sets of data, and any set of the multiple sets of data includes picture data and labels that annotate table structure features; The image is processed through the image text layout extraction model to obtain the image text layout information, wherein the image text layout extraction model is generated by training a convolutional neural network through multiple sets of data, and any set of the multiple sets of data includes image data and labels that mark the text layout position.

5. The system according to claim 1, wherein: According to the text attribute set and the format layout feature set, a check binary tree is constructed, and the execution steps of the check tree construction module include: The format layout feature set includes a table structure feature set and a text attribute layout position set; According to the text attribute set, construct a first check binary subtree; Constructing a second check binary subtree according to the table structure feature set; According to the text attribute layout position set, a third check binary subtree is constructed; Add the first check binary subtree, the second check binary subtree and the third check binary subtree into the check binary tree.

6. The system according to claim 5, characterized in that According to the text attribute set, a first check binary subtree is constructed, and the execution steps of the check tree construction module also include: Extract the first text attribute of the text attribute set and set it as the root node of the binary tree; Extract the second text attribute of the text attribute set and set it as a secondary node of the binary tree; Until the Nth text attribute of the text attribute set is extracted, it is set as the Nth level node of the binary tree; Connect the binary tree root node, the binary tree secondary node until the binary tree N-level node to generate a first check binary subtree.

7. The system according to claim 1, characterized in that Accessing multiple open databases according to multiple target qualification types, and performing information comparison on the multiple binary trees of work orders to be verified respectively, the execution steps of the qualification verification module include: Accessing a first open database according to the first target qualification type, and extracting a first binary tree of work orders to be verified; Obtain the root node data of the first work order to be verified binary tree; When the root node data is a layout feature, the qualification information to be verified and the format layout feature stored in the root node are compared. If they are consistent, the root node data is considered to be consistent. If they are inconsistent, a qualification verification failure message is sent to the corresponding mobile terminal; When the root node data is text, the text is searched in the first open database. If the text is found, it is considered that the root node data matches. If the text is not found, a qualification verification failure message is sent to the corresponding mobile terminal. When the root node data matches, obtain the secondary node data of the first work order binary tree to be verified and continue verification; Until every node of the binary tree of the first work order to be verified is verified to be consistent, it is considered to be consistent.

8. The system of claim 1, wherein: RPA robots also include: The visualization end is used to display the user and work order verification status in the visualization interface when the binary tree of multiple work orders to be verified is completed. The work order verification status includes not started, in progress, and completed.

Images