Data processing method and device, equipment, medium and program product

Abstract

The invention provides a data processing method and device, equipment, a storage medium and a program product, and can be applied to the technical field of artificial intelligence and the technical field of financial science and technology. The data processing method comprises the following steps: performing anti-counterfeiting identification on a check to be processed based on three-dimensional imaging and / or multispectral imaging, and determining the check to be processed as a real check; under the condition that the check to be processed is a real check, extracting text information from the check to be processed; and mapping the text information to a preset incoming bill template to obtain a target incoming bill corresponding to the check to be processed.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and more specifically to a data processing method, apparatus, device, medium, and program product. Background Art

[0002] A transfer check is a document issued by the drawee, notifying the bank to withdraw funds from their account. A bank deposit slip is a special document used by bank account holders to deposit the check. When using a transfer check to make a transfer, customers must complete a bank deposit slip and use it alongside the check to ensure the funds are accurately transferred to the designated account.

[0003] Currently, when customers visit bank branches to conduct transactions, they must manually fill out check receipts, a cumbersome process. Furthermore, manual receipts often come with issues such as irregularities, unclear handwriting, and typos, resulting in low accuracy and efficiency in receipt generation. Summary of the Invention

[0004] In view of the above problems, the present application provides a data processing method, apparatus, device, medium and program product for improving the security and efficiency of bill generation.

[0005] According to the first aspect of the present application, a data processing method is provided, including: performing anti-counterfeiting identification on a check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a genuine check; when the check to be processed is a genuine check, extracting text information from the check to be processed; mapping the text information to a preset invoice template to obtain a target invoice corresponding to the check to be processed.

[0006] According to an embodiment of the present application, anti-counterfeiting identification is performed on the check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a real check, including: constructing a three-dimensional model of the check to be processed based on the depth data and two-dimensional image data of the check to be processed; and analyzing the physical characteristics of the three-dimensional model to obtain a first identification result of the check to be processed.

[0007] According to an embodiment of the present application, anti-counterfeiting identification is performed on a check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a genuine check, including: irradiating the check to be processed based on at least one light source to obtain image information of the check to be processed under different spectra; extracting characteristic parameters of each area in the check to be processed under different spectra from the multispectral image information; inputting the characteristic parameters as input data into a pre-trained analysis model, and the analysis model outputting a second recognition result of the check to be processed based on the input data.

[0008] According to an embodiment of the present application, when the check to be processed is a real check, target information is extracted from the check to be processed, including: obtaining image data of the check to be processed; performing layout analysis on the image data to determine the position and type of different information areas in the image data; based on the position and type of each information area, obtaining multiple similarity transformation images corresponding to the image data, wherein the similarity transformation image only contains a specific information area; and separately identifying the multiple similarity transformation images to extract text information from the image data.

[0009] According to an embodiment of the present application, the types of similarity transformation images include a first-class similarity transformation image and a second-class similarity transformation image, and multiple similarity transformation images are respectively identified to extract text information in the image data. It also includes: identifying and verifying the target features in the first-class similarity transformation image to determine whether the check is a valid check, wherein the target feature is at least one of a seal or a signature; and identifying the characters in the second-class similarity transformation image to obtain the text information in the image data.

[0010] According to an embodiment of the present application, the target information is mapped to a preset receipt template to obtain a target receipt corresponding to the check to be processed, including: mapping the identified text information with the fields in the target receipt template according to preset rules to obtain a field mapping result; performing semantic recognition on the field mapping result, and determining the field attribution of the text information based on the semantic recognition result and the field mapping result; filling the text information into the corresponding field in the target receipt template according to the field attribution of the text information to obtain the target receipt corresponding to the check to be processed.

[0011] According to an embodiment of the present application, semantic recognition is performed on the field mapping results, and the field ownership of the text information is determined based on the semantic recognition results and the field mapping results, including: checking the field mapping results to determine whether there is ambiguity in the text information; wherein the ambiguity is that the text information may be related to multiple different fields; performing semantic analysis on the ambiguous field mapping; adjusting the ambiguous field mapping based on the semantic analysis results to determine the field ownership of the text information.

[0012] According to an embodiment of the present application, the target invoice template is determined from a plurality of preset invoice templates in at least one of the following ways: performing template matching based on the position and type of different information areas in the image data, and determining the target invoice template corresponding to the pending check from a plurality of preset invoice templates; determining the target invoice template corresponding to the pending check from a plurality of preset invoice templates based on the user habits of the operating user.

[0013] The second aspect of the present application provides a data processing device, including: an identification module, used to perform anti-counterfeiting identification on the pending check based on three-dimensional imaging and / or multispectral imaging, and determine that the pending check is a genuine check; an extraction module, used to extract text information from the pending check when the pending check is a genuine check; and a mapping module, used to map the text information to a preset invoice template to obtain a target invoice corresponding to the pending check.

[0014] The third aspect of the present application provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0015] The fourth aspect of the present application further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the computer program or instructions are executed by a processor.

[0016] The fifth aspect of the present application further provides a computer program product, comprising a computer program or instructions, which implement the steps of the above method when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above contents and other objects, features and advantages of the present application will become more apparent through the following description of the embodiments of the present application with reference to the accompanying drawings, in which:

[0018] Figure 1 Schematically illustrates an application scenario diagram of a data processing method, apparatus, device, medium, and program product according to an embodiment of the present application;

[0019] Figure 2 The following schematically shows a flow chart of a data processing method according to an embodiment of the present application;

[0020] Figure 3 A flowchart schematically illustrates anti-counterfeiting identification of a check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a genuine check according to an embodiment of the present application;

[0021] Figure 4 A flowchart schematically illustrates anti-counterfeiting identification of a check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a genuine check according to an embodiment of the present application;

[0022] Figure 5 A flowchart schematically illustrates a method for extracting target information from a check to be processed when the check to be processed is a genuine check according to an embodiment of the present application;

[0023] Figure 6A flowchart schematically illustrates how to identify multiple similarity-transformed images and extract text information from image data according to an embodiment of the present application;

[0024] Figure 7 A flowchart schematically illustrates mapping target information to a preset invoice template to obtain a target invoice corresponding to a check to be processed according to an embodiment of the present application;

[0025] Figure 8 A block diagram schematically illustrates a structure of a data processing device according to an embodiment of the present application; and

[0026] Figure 9 A block diagram of an electronic device suitable for implementing a data processing method according to an embodiment of the present application is schematically shown. DETAILED DESCRIPTION

[0027] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present application. In the detailed description below, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present application. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessarily confusing the concepts of the present application.

[0028] The terms used herein are only for describing specific embodiments and are not intended to limit this application. The terms "comprise," "include," etc. used herein indicate the presence of the features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0029] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0030] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0031] It should be noted that the data processing method and device of the present application can be used in the field of financial technology, and can also be used in any field other than the field of financial technology. The application field of the data processing method and device of the present application is not limited.

[0032] In the technical solution of this application, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0033] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided in the embodiments of the present application all provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0034] An embodiment of the present application provides a data processing method, including: performing anti-counterfeiting identification on a check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a genuine check; if the check to be processed is a genuine check, extracting text information from the check to be processed; mapping the text information to a preset invoice template to obtain a target invoice corresponding to the check to be processed.

[0035] Figure 1 The application scenario diagram of the data processing method, apparatus, device, medium and program product according to the embodiments of the present application is schematically shown.

[0036] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or optical fiber cables.

[0037] A user may use a first terminal device 101, a second terminal device 102, or a third terminal device 103 to interact with a server 105 via a network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, or the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (for example only).

[0038] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0039] The server 105 may be a server that provides various services, such as a background management server (for example only) that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process received data such as user requests, and feed back processing results (e.g., web pages, information, or data obtained or generated based on user requests) to the terminal devices.

[0040] It should be noted that the data processing method provided in the embodiment of the present application can generally be executed by the server 105. Accordingly, the data processing device provided in the embodiment of the present application can generally be set in the server 105. The data processing method provided in the embodiment of the present application can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the data processing device provided in the embodiment of the present application can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0041] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0042] The following will be based on Figure 1 The scene described by Figures 2 to 7 The data processing method according to the embodiment of the present application is described in detail.

[0043] Figure 2 The flowchart of the data processing method according to the embodiment of the present application is schematically shown.

[0044] like Figure 2 As shown, the data processing method of this embodiment includes operations S210 to S230.

[0045] In operation S210 , anti-counterfeiting identification is performed on the check to be processed based on three-dimensional imaging and / or multi-spectral imaging to determine whether the check to be processed is a genuine check.

[0046] In some embodiments, in response to detecting that a check to be processed is placed in the scanning area of ​​the device, the three-dimensional imaging module and the multi-spectral imaging module are controlled to perform anti-counterfeiting identification on the check to be processed respectively, and the authenticity of the check to be processed is comprehensively judged based on the results of the three-dimensional feature analysis and the multi-spectral feature analysis. When the three-dimensional features and multi-spectral features of the check to be processed both meet the features of a real check, the check to be processed is determined to be a real check.

[0047] In some embodiments, 3D imaging compares a generated 3D model of a check to be processed with a pre-stored 3D model of an authentic check to determine whether the 3D features of the check to be processed are identical to those of the authentic check. Check ink, security features, and other components exhibit different characteristics under different light spectra. Multispectral imaging can capture image information of the check under different light spectra and determine its authenticity by analyzing the security features presented under these different light spectra.

[0048] In operation S220 , if the check to be processed is a genuine check, text information is extracted from the check to be processed.

[0049] In some embodiments, the image of the pending check that has passed the anti-counterfeiting identification is pre-processed. For example, after the pending check passes the anti-counterfeiting identification, image data of the pending check is captured using an image acquisition module, and the captured image data is pre-processed. Exemplary pre-processing operations may include denoising, contrast enhancement, binarization, and the like.

[0050] In some embodiments, text recognition and text extraction are performed on the pre-processed image data to obtain text information in the check to be processed.

[0051] In operation S230, the text information is mapped to a preset invoice template to obtain a target invoice corresponding to the check to be processed.

[0052] In some embodiments, an appropriate invoice template is selected from multiple pre-defined templates based on the check type and business requirements. The templates define the location and format of each field. The extracted text information is mapped to the fields in the invoice template, and the information is then populated into the corresponding fields to generate a target invoice corresponding to the check to be processed. For example, the drawee's name on the check is populated into the "Payer" field of the invoice.

[0053] The data processing method provided in this application uses multiple methods to perform anti-counterfeiting identification on pending checks, effectively improving the accuracy of anti-counterfeiting identification and enhancing the security of data processing. After a pending check passes anti-counterfeiting identification, text information is extracted from the check and automatically filled in and formatted according to a preset template, enabling automated generation of invoices. This ensures the standardization and consistency of invoices, effectively improving invoice generation efficiency, and reducing manual intervention, labor costs, and potential errors associated with manual data entry.

[0054] Figure 3 The flowchart schematically shows a method for performing anti-counterfeiting identification on a check to be processed based on three-dimensional imaging and / or multi-spectral imaging to determine whether the check to be processed is a genuine check according to an embodiment of the present application.

[0055] like Figure 3 As shown, in the embodiment of the present application, anti-counterfeiting identification of a check to be processed is performed based on three-dimensional imaging and / or multi-spectral imaging, and determining that the check to be processed is a genuine check includes operations S310 to S320.

[0056] In operation S310 , a three-dimensional model of the check to be processed is constructed based on the depth data and the two-dimensional image data of the check to be processed.

[0057] In some embodiments, a depth sensor may be used to scan the check to be processed to obtain depth data of the check to be processed, and a common camera may be used to photograph the check to be processed to obtain a two-dimensional image of the check under visible light.

[0058] Exemplary depth sensors include structured light sensors and time-of-flight sensors. Structured light sensors transmit a specific light pattern (such as stripes or grids) onto the check surface. A camera then captures the deformation of the light pattern reflected from the check surface. Using triangulation, the sensor calculates the distance from each point on the check surface to obtain depth data. Time-of-flight sensors calculate distance and obtain depth data by measuring the time difference between the transmitted light pulse and the received reflected light pulse.

[0059] In some embodiments, the collected depth data is preprocessed and point cloud data of the check surface is generated based on the depth data. A point cloud is a collection of points in a three-dimensional space, each containing position information (X, Y, and Z coordinates). The point cloud data is then registered with the two-dimensional image data, essentially finding a correspondence between the point cloud data and the two-dimensional image data so that they are spatially aligned. Using the registered point cloud data and the two-dimensional image data, a three-dimensional model of the check is constructed using a surface reconstruction algorithm (such as Poisson reconstruction or marching cubes).

[0060] In operation S320 , the physical features of the three-dimensional model are analyzed to obtain a first recognition result of the check to be processed.

[0061] In some embodiments, physical features of the check are extracted from the three-dimensional model and analyzed to obtain a first recognition result of the check to be processed, wherein the first recognition result is used to characterize the authenticity result of the check to be processed obtained after three-dimensional imaging.

[0062] For example, physical features may include check thickness, material, texture, etc. A comprehensive judgment is made on multiple physical features in the three-dimensional model. If all features are consistent with the features of a real check, the check to be processed is considered to be a real check. If one or more features do not meet the requirements, the check to be processed may be considered to be forged, tampered with, or other problems.

[0063] For example, the thickness of different areas of a check can be calculated to determine whether there is localized thickening or thinning. Material recognition algorithms can be used to analyze the composition and texture of the check paper and compare it with the material characteristics of an authentic check.

[0064] For example, a counterfeit check might be tampered with through patching or alteration, which can cause the thickness of a localized area of ​​the check to change. Different banks and different types of checks typically use paper with unique composition and texture. If the material of the check being processed does not match that of an authentic check sample, it may be a counterfeit check.

[0065] This application uses 3D imaging to construct a 3D model of the check to be identified. This 3D model incorporates the check's depth information, providing a more comprehensive reflection of the check's physical characteristics. This effectively improves the accuracy and reliability of the anti-counterfeiting identification results for the check being processed, increasing the recognition rate of counterfeit checks and reducing the risk of false positives and missed detections. For different check types, the corresponding feature analysis parameters and recognition rules can be adjusted based on the check's characteristics, effectively enhancing the flexibility and scalability of anti-counterfeiting identification.

[0066] Figure 4 The flowchart schematically shows a method for performing anti-counterfeiting identification on a check to be processed based on three-dimensional imaging and / or multi-spectral imaging to determine whether the check to be processed is a genuine check according to an embodiment of the present application.

[0067] like Figure 4 As shown, in the embodiment of the present application, anti-counterfeiting identification of a check to be processed is performed based on three-dimensional imaging and / or multi-spectral imaging, and determining that the check to be processed is a genuine check includes operations S410 to S430.

[0068] In operation S410 , the check to be processed is illuminated by at least one light source to obtain image information of the check to be processed under different spectrums.

[0069] In some embodiments, at least one suitable light source is selected to illuminate the check being processed. Examples include ultraviolet, infrared, and visible light sources. These light sources can be used individually or in combination. Different light sources emit light with different wavelengths, stimulating the properties of different substances on the check. For example, ultraviolet light can stimulate the luminescence of fluorescent substances on the check, while infrared light can penetrate paper of a certain thickness to obtain information within the paper. Light sources are positioned at appropriate angles and distances around the check being processed to ensure uniform illumination of the check surface, avoiding shadows or uneven illumination.

[0070] In some embodiments, a multispectral camera is used to capture images of the illuminated check to be processed to obtain image information of the check to be processed under different spectra. The multispectral camera can simultaneously obtain image information of multiple spectral bands.

[0071] In operation S420 , characteristic parameters of each region of the check to be processed under different spectra are extracted from the multispectral image information.

[0072] In some embodiments, the collected multispectral images are subjected to denoising and registration processing, and the check is divided into different areas, such as the ticket number area, the amount area, the seal area, and the anti-counterfeiting mark area, based on the structure of the check and the distribution of anti-counterfeiting features. The area division can be based on the standard check template or actual business needs. The characteristic parameters under different spectra may include, for example, reflectivity, absorptivity, etc. These characteristic parameters can reflect the chemical composition and physical structure of the substances on the check.

[0073] In operation S430 , the feature parameters are input as input data to the pre-trained analysis model, and the analysis model outputs a second recognition result of the check to be processed based on the input data.

[0074] In some implementations, spectral signatures are obtained by analyzing the reflectivity and transmittance of each region under different light spectra. Authentic checks exhibit specific spectral responses under different light spectra. By comparing these spectral signatures with pre-stored spectral signatures of authentic checks, the authenticity of the check can be determined.

[0075] Exemplarily, the analysis model is trained using a large amount of multispectral image data from genuine and counterfeit checks. The analysis model can be, for example, a neural network model or a support vector machine model. Feature parameters are input to the analysis model as input data. The analysis model then determines the authenticity of the check based on the check's multispectral feature parameters and outputs a second recognition result indicating the authenticity of the check.

[0076] This application uses multispectral imaging technology to perform anti-counterfeiting identification on pending checks, thereby improving the accuracy of anti-counterfeiting identification. On the one hand, the inks used by different banks and for different types of checks have specific spectral characteristics. Multispectral imaging technology can detect the spectral reflectance curve of the ink on the check and compare it with the spectral characteristics of authentic ink. If the spectral characteristics of the ink do not match those of authentic ink, the check may be counterfeit. On the other hand, many anti-counterfeiting markings on checks (such as fluorescent fibers and color-changing inks) exhibit unique visual effects under different spectra. Multispectral imaging technology can clearly capture the characteristics of these anti-counterfeiting markings, such as the distribution of fluorescent fibers and the color changes of color-changing inks, effectively identifying counterfeit checks.

[0077] According to the embodiments of the present application, Figure 2 The data processing method shown further includes the following operation: determining the anti-counterfeiting recognition result of the check to be processed based on the first recognition result and the second recognition result.

[0078] In some embodiments, for example, the anti-counterfeiting identification result of the check to be processed can be determined by a weighted calculation method or a logic method.

[0079] For example, a weighted calculation method can be used: different weights are assigned to the first and second recognition results based on the accuracy and reliability of 3D imaging and multispectral imaging authentication. For example, if 3D imaging has a higher accuracy rate in past tests, a higher weight is assigned to it; conversely, if multispectral imaging has a higher accuracy rate, a higher weight is assigned to it. The two recognition results are then weighted together to obtain the final recognition result.

[0080] Exemplarily, the logic method may be: if both the first recognition result and the second recognition result are "authentic", the check to be processed is determined to be an authentic check; if only one of the recognition results is "forged", the check is determined to be a forged check.

[0081] Determining the anti-counterfeiting identification results of pending checks based on the respective recognition results of 3D imaging and multispectral imaging can effectively improve anti-counterfeiting identification accuracy. 3D imaging technology can detect features such as the check's geometry and surface texture, while multispectral imaging technology can analyze the check's optical properties under different light spectra. By jointly determining the identification results of pending checks based on the respective recognition results of both technologies, counterfeit features that a single technology might overlook can be accurately detected, improving overall identification accuracy. Furthermore, employing multiple identification methods effectively improves the fault tolerance of data processing. If one identification technology experiences an error due to some reason (such as equipment failure or environmental interference), the other technology can still provide accurate identification results, thus ensuring the reliability of the entire data processing.

[0082] Figure 5The flowchart of extracting target information from a check to be processed is schematically shown in accordance with an embodiment of the present application when the check to be processed is a genuine check.

[0083] like Figure 5 As shown, in the embodiment of the present application, when the check to be processed is a real check, extracting target information from the check to be processed includes operations S510 to S540.

[0084] In operation S510 , image data of a check to be processed is acquired.

[0085] In some embodiments, based on actual business scenarios and needs, appropriate image acquisition equipment, such as a high-resolution scanner, a professional camera, etc., is selected to obtain clear and complete check image data. The image data may also be the two-dimensional image data acquired in operation S310.

[0086] In operation S520, the image data is format parsed to determine the positions and types of different information areas in the image data.

[0087] In some embodiments, the acquired image data is preprocessed, and layout features are extracted from the preprocessed image data. These layout features may include, for example, check edges, lines, stamps, and text areas. Edge detection algorithms (such as the Canny algorithm) or line detection algorithms (such as the Hough transform) can be used to extract these layout features. Based on the extracted layout features, template matching, rule matching, or machine learning algorithms (such as convolutional neural networks) are used to analyze the image data and determine the location and type of different information areas. For example, the check number area, amount area, date area, and stamp area can be identified, and the coordinate range and type information of each area in the image can be recorded.

[0088] In operation S530 , a plurality of similarity transformation images corresponding to the image data are acquired based on the position and type of each information region, wherein the similarity transformation images only include a certain specific information region.

[0089] In some embodiments, the original image data is cropped based on the location and type of different information regions obtained through format analysis to generate a similarity-transformed image containing only a specific information region. For example, if the text information in the ticket number region needs to be extracted, the image portion corresponding to the ticket number region is cropped. The generated similarity-transformed images can be saved to a designated folder or database, and corresponding identification information, such as the check number and region type, is added to each similarity-transformed image for subsequent processing and recognition.

[0090] In operation S540 , the plurality of similar transformed images are respectively recognized to extract text information from the image data.

[0091] In some embodiments, each similarity-transformed image undergoes further preprocessing and optimization to improve text recognition accuracy. For example, preprocessing can include image enhancement operations such as contrast adjustment and sharpening to make text more legible. Optical character recognition (OCR) technology can be used to perform text recognition on similarity-transformed images, converting the text in the similarity-transformed image into an editable text format.

[0092] Exemplarily, text recognition can be performed on multiple similarity transformation images in parallel, thereby improving the efficiency of text recognition. For example, different similarity transformation images can be assigned to different computing nodes for recognition, effectively shortening the overall processing time.

[0093] In some embodiments, the text recognition results of the respective similar transformed images are integrated to obtain the complete check text information.

[0094] This application focuses the recognition task on a smaller range by generating a similarity transformation image that only contains a specific information area, reducing background interference and the influence of other irrelevant information, thereby improving the accuracy of text recognition. For example, when identifying the ticket number area, only the image of the ticket number area is processed, avoiding interference from text or patterns in other areas. This recognition method does not rely on a specific check template. Through layout analysis, it can automatically identify the location and type of information areas of different checks and generate corresponding similarity transformation images. Therefore, it can adapt to checks of various types and layouts and has strong versatility and flexibility.

[0095] Figure 6 The flowchart of respectively identifying a plurality of similar transformed images and extracting text information from the image data according to an embodiment of the present application is schematically shown.

[0096] like Figure 6 As shown, the embodiment of the present application recognizes a plurality of similar transformed images respectively, and extracts text information from the image data, including operations S610 to S620.

[0097] In operation S610 , target features in a class of similarity transformation images are identified and verified to determine whether the check is a valid check, wherein the target feature is at least one of a seal or a signature.

[0098] In operation S620 , characters in the two types of similarity transformation images are recognized to obtain text information in the image data.

[0099] In some embodiments, the similarity transformation image may include a first-type similarity transformation image and a second-type similarity transformation image. The first-type similarity transformation image mainly includes at least one key position information of a seal or signature on a check, and the second-type similarity transformation image mainly includes characters on the check.

[0100] In some implementations, image processing technology can be used to extract features of a seal or signature. For a seal, features such as the shape, edge contour, and texture of the seal can be extracted; for a signature, features such as the stroke trend, tilt angle, and connected stroke features of the signature can be extracted. The extracted features are compared with the features of a real seal or signature pre-stored in a feature database. For example, a similarity measurement method such as Euclidean distance, cosine similarity, etc. can be used to calculate the similarity between the feature to be identified and the feature in the database. If the similarity between the target feature and a feature in the database is higher than the threshold, the seal or signature is determined to be valid, that is, the check is a valid check. Otherwise, it is determined to be invalid, that is, the check is an invalid check. Exemplarily, when it is determined that the check is a valid check, operation S620 is performed.

[0101] In some embodiments, an OCR engine or a trained character recognition model can be used to recognize the segmented characters. The character image is input into the model, and the model outputs the corresponding character text. The recognized characters are then integrated into complete text information, such as ticket number, amount, date, etc.

[0102] This application specifically identifies and verifies seals or signatures within a first-class similarity-transformed image, accurately determining the authenticity of key anti-counterfeiting features on checks and improving data processing security and accuracy. Furthermore, character recognition within a second-class similarity-transformed image containing only characters effectively avoids interference from irrelevant information and improves character recognition accuracy.

[0103] Figure 7 The flowchart of mapping target information to a preset deposit slip template to obtain a target deposit slip corresponding to a check to be processed is schematically shown according to an embodiment of the present application.

[0104] like Figure 7 As shown, in the embodiment of the present application, mapping the target information to the preset deposit slip template to obtain the target deposit slip corresponding to the to-be-processed check includes operations S710 to S730.

[0105] In operation S710, the recognized text information is mapped to fields in the target invoice template according to a preset rule to obtain a field mapping result.

[0106] In some embodiments, the preset rules may include keyword matching rules, field order rules, coordinate positioning rules, etc. For example, information such as "payee", "payer", "amount", "purpose", and "remarks" are used as preset keywords, and these keywords are searched in the recognized text information. If the match is successful, the text content near the keyword is bound to the corresponding template field. Alternatively, the field order of some invoices is relatively fixed, and the recognized text can be filled in the response field of the template in sequence according to the preset field order rules. Alternatively, for some checks with relatively standardized formats, the position of a specific field can be located by a pre-set coordinate range. Based on these coordinate ranges, the system extracts the text in the corresponding area from the OCR recognition results and fills it into the corresponding field of the template. For example, the coordinate range of the "payee" area is pre-set, and when the system recognizes the image, it only extracts the text in this area as the content of the "payee" field.

[0107] In operation S720 , semantic recognition is performed on the field mapping result, and the field attribution of the text information is determined based on the semantic recognition result and the field mapping result.

[0108] In some embodiments, natural language processing (NLP) technology is used to perform semantic recognition on the field mapping results. Methods such as lexical analysis, syntactic analysis, and semantic role labeling can be used to understand the semantic content of text information. Pre-trained language models (such as BERT and GPT) can also be used to semantically understand and classify text information to further clarify its meaning. The semantic recognition results are combined with the field mapping results to determine the final field assignment of the text information.

[0109] In some embodiments, operation S720 may, for example, include: checking the field mapping results to determine whether there is ambiguity in the text information; wherein the ambiguity is that the text information may be related to multiple different fields; performing semantic analysis on the ambiguous field mapping; adjusting the ambiguous field mapping according to the semantic analysis results to determine the field affiliation of the text information.

[0110] For example, when multiple paragraphs of text appear in the recognition results, and these texts may be related to the "Purpose" or "Remarks" fields, these situations are marked as ambiguous. For ambiguous field mappings, natural language processing (NLP) technology can be used for semantic analysis. NLP technology can perform operations such as part-of-speech tagging, syntactic analysis, and semantic role tagging on text to understand the meaning of the text. Combined with contextual logic, the ownership of information is determined. For example, if a text such as "Purpose: Payment" is recognized, semantic analysis can understand that the content after "Purpose" is a specific description of the purpose, so "Payment" is automatically classified into the "Purpose" field of the invoice; if "Remarks: Expedited" is recognized, the system will also classify "Expedited" into the "Remarks" field based on semantics and contextual logic.

[0111] In operation S730, the text information is filled into corresponding fields in the target invoice template according to the field attribution of the text information, and the target invoice corresponding to the to-be-processed check is obtained.

[0112] In some embodiments, before entering text into the target invoice template, the text can be formatted according to the template field's formatting requirements. For example, if a date field in the template requires the format "YYYY / MM / DD" and the recognized date is "2023-10-01," it needs to be converted to "2023 / 10 / 01." Numerical information, such as amounts, also requires appropriate formatting, such as adding a currency symbol and retaining an appropriate number of decimal places.

[0113] In some embodiments, the converted text information is accurately filled into the corresponding fields of the target invoice template according to the field attribution of the text information. After all the text information is filled in, the target invoice corresponding to the pending check is generated.

[0114] This application uses preset rules and semantic recognition to jointly determine the field ownership of text information, thereby realizing the automatic filling of text information and the automatic generation of target invoices. Among them, the preset rules can be customized according to the characteristics of the invoice template, and can more accurately match the identified text information with the template fields, thereby improving the accuracy of field mapping. Semantic recognition can further understand the meaning of text information, resolve some fuzzy matching and ambiguous issues, and ensure that text information is accurately filled in the corresponding fields. Combining preset rules with semantic recognition can effectively improve the efficiency and accuracy of target invoice generation.

[0115] In some embodiments, the target invoice template may be determined by at least one of the following methods: performing template matching based on the location and type of different information areas in the image data to determine the target invoice template corresponding to the pending check from multiple preset invoice templates; and / or determining the target invoice template corresponding to the pending check from multiple preset invoice templates based on the user habits of the operating user.

[0116] For example, feature description information can be established for each preset invoice template. This feature description information can include, for example, the coordinate positions of each information area within the template, the relative relationships between the areas, and the text features corresponding to each area. The position and type of the different information areas in the image data of the check to be processed are matched with the feature description information of each invoice template, and the target invoice template corresponding to the check to be processed is determined based on the matching results. A similarity measurement method (such as Euclidean distance or cosine similarity) can be used to calculate the similarity between the features of each information area in the image data and the features of each preset invoice template. A similarity threshold is set. If the similarity of a template exceeds the threshold, the template is considered to match the check image and is determined as the target invoice template. If multiple templates have similarities above the threshold, the template with the highest similarity is selected as the target template.

[0117] Performing template matching based on the location and type of information areas in image data can effectively improve the accuracy of template matching and increase data processing efficiency.

[0118] For example, the identity information of the user performing the current operation (e.g., user ID, user name, etc.) can be obtained. Based on the user's user habits, a target invoice template corresponding to the pending check can be determined from multiple preset invoice templates. User habits are obtained by collecting historical data on the user's selection of invoice templates. For example, this data may include information such as the time of selection, check type, and invoice type. By analyzing this collected user habit data, the frequency with which users select various invoice templates in different scenarios (e.g., different check types, different transaction times, etc.) can be calculated. For example, if statistics show that a user selects Template A 80% of the time when processing corporate checks, it can be assumed that the user prefers Template A in these scenarios. Based on information such as the current check type (e.g., corporate check, personal check, etc.), transaction time, and user habit data, the invoice template most likely to be used by the user in the current scenario can be selected as the target invoice template.

[0119] Template matching based on user habits provides personalized service. Automatically recommending appropriate invoice templates based on user habits enhances the user experience. For users who frequently use a specific template, the system can directly recommend the target template, reducing the time spent manually selecting templates and improving business processing efficiency.

[0120] According to an embodiment of the present application, the data processing method provided in the embodiment of the present application further includes:

[0121] After the target invoice for the pending check is generated, the transaction information will be automatically cleared.

[0122] In some embodiments, the current transaction information is cleared in response to receiving a specific signal. For example, after the user confirms that the target invoice is correct, the target invoice is printed via the printing module. After successful printing, a "print successful" signal is received from the printing module. In response to receiving this signal, the current transaction process is confirmed to have ended, and the current transaction information is automatically cleared.

[0123] In some embodiments, a timeout mechanism may be provided. If a user makes an erroneous operation within a specific time range, the current business process is automatically determined to have ended, and the business information is cleared.

[0124] By automatically clearing business information, data security and privacy are effectively protected, and the risk of information leakage is reduced. Automatically clearing business information for completed transactions can free up storage space in a timely manner, preventing it from being filled with invalid data. This ensures that the system has enough space to process new transactions, improving system efficiency and stability.

[0125] Based on the above data processing method, this application also provides a data processing device. Figure 8 The device is described in detail.

[0126] Figure 8 The structural block diagram of a data processing device according to an embodiment of the present application is schematically shown.

[0127] like Figure 8 As shown, the data processing device 800 of this embodiment includes an identification module 810 , an extraction module 820 and a mapping module 830 .

[0128] The identification module 810 is used to perform anti-counterfeiting identification on the check to be processed based on three-dimensional imaging and / or multispectral imaging to determine whether the check to be processed is a genuine check. In one embodiment, the identification module 810 can be used to perform the operation S210 described above, which will not be repeated here.

[0129] The extraction module 820 is used to extract text information from the check to be processed when the check to be processed is a genuine check. In one embodiment, the extraction module 820 can be used to perform the operation S220 described above, which will not be repeated here.

[0130] The mapping module 830 is used to map the text information to a preset invoice template to obtain a target invoice corresponding to the check to be processed. In one embodiment, the mapping module 830 can be used to perform the operation S230 described above, which will not be repeated here.

[0131] According to embodiments of the present application, any multiple modules among the identification module 810, extraction module 820, and mapping module 830 may be combined into a single module, or any one of these modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module. According to embodiments of the present application, at least one of the identification module 810, extraction module 820, and mapping module 830 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or may be implemented in hardware or firmware through any other reasonable means of circuit integration or packaging, or may be implemented in any one of the three implementation methods of software, hardware, and firmware, or any appropriate combination of any of these. Alternatively, at least one of the identification module 810, extraction module 820, and mapping module 830 may be at least partially implemented as a computer program module that, when executed, performs the corresponding functionality.

[0132] Figure 9 A block diagram of an electronic device suitable for implementing a data processing method according to an embodiment of the present application is schematically shown.

[0133] like Figure 9 As shown, an electronic device 900 according to an embodiment of the present application includes a processor 901, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 902 or programs loaded from a storage unit 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or related chipsets and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to the embodiment of the present application.

[0134] Various programs and data required for the operation of the electronic device 900 are stored in the RAM 903. The processor 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. The processor 901 performs various operations of the method flow according to the embodiment of the present application by executing the programs in the ROM 902 and / or the RAM 903. It should be noted that the programs may also be stored in one or more memories other than the ROM 902 and the RAM 903. The processor 901 may also perform various operations of the method flow according to the embodiment of the present application by executing the programs stored in the one or more memories.

[0135] According to an embodiment of the present application, electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to bus 904. Electronic device 900 may also include one or more of the following components connected to I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage section 908 including a hard disk; and a communication section 909 including a network interface card such as a LAN card or modem. Communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to I / O interface 905 as needed. Removable media 911, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 910 as needed, so that computer programs read from the removable media can be installed into storage section 908 as needed.

[0136] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not be incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiments of this application is implemented.

[0137] According to an embodiment of the present application, a computer-readable storage medium may be a non-volatile computer-readable storage medium, and may include, for example, but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present application, a computer-readable storage medium may include the ROM 902 and / or RAM 903 described above and / or one or more memories other than ROM 902 and RAM 903.

[0138] The embodiments of the present application also include a computer program product, which includes a computer program containing program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the data processing method provided in the embodiments of the present application.

[0139] The computer program executes the above functions defined in the system / device of the embodiment of the present application when the processor 901 executes the computer program. According to the embodiment of the present application, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0140] In one embodiment, the computer program may be stored on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may be transmitted and distributed in the form of a signal on a network medium, downloaded and installed via the communication portion 909, and / or installed from a removable medium 911. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0141] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from a removable medium 911. When the computer program is executed by the processor 901, the above-mentioned functions defined in the system of the embodiment of the present application are performed. According to the embodiment of the present application, the systems, devices, means, modules, units, etc. described above can be implemented by computer program modules.

[0142] According to an embodiment of the present application, the program code for executing the computer program provided by the embodiment of the present application can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0143] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of the boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0144] Those skilled in the art will appreciate that the features described in the various embodiments of this application may be combined and / or coupled in various ways, even if such combinations or couplings are not explicitly described in this application. In particular, the features described in the various embodiments of this application may be combined and / or coupled in various ways without departing from the spirit and teachings of this application. All such combinations and / or couplings fall within the scope of this application.

Claims

1. A data processing method, characterized in that: The method comprises: performing anti-counterfeiting identification on the check to be processed based on three-dimensional imaging and / or multispectral imaging to determine that the check to be processed is a genuine check; extracting text information from the check to be processed if the check to be processed is a genuine check; The text information is mapped to a preset invoice template to obtain a target invoice corresponding to the check to be processed.

2. The data processing method according to claim 1, wherein: The anti-counterfeiting identification of the check to be processed based on three-dimensional imaging and / or multi-spectral imaging to determine that the check to be processed is a genuine check includes: constructing a three-dimensional model of the check to be processed based on the depth data and the two-dimensional image data of the check to be processed; The physical features of the three-dimensional model are analyzed to obtain a first recognition result of the check to be processed.

3. The data processing method according to claim 1, wherein: The anti-counterfeiting identification of the check to be processed based on three-dimensional imaging and / or multi-spectral imaging to determine that the check to be processed is a genuine check includes: irradiating the check to be processed based on at least one light source to obtain image information of the check to be processed under different spectra; extracting characteristic parameters of each region of the check to be processed under different spectra from the multispectral image information; The feature parameters are input as input data to a pre-trained analysis model, and the analysis model outputs a second recognition result of the check to be processed based on the input data.

4. The data processing method according to claim 1, wherein: When the check to be processed is a genuine check, extracting target information from the check to be processed includes: Acquiring image data of the check to be processed; Performing format analysis on the image data to determine the positions and types of different information areas in the image data; Based on the position and type of each information region, a plurality of similarity transformation images corresponding to the image data are obtained, wherein the similarity transformation images only contain a certain specific information region; The plurality of similar transformed images are respectively recognized to extract text information from the image data.

5. The data processing method according to claim 4, characterized in that: The types of the similarity transformation images include a first-class similarity transformation image and a second-class similarity transformation image, and the steps of respectively identifying the plurality of similarity transformation images and extracting text information from the image data further include: Identifying and verifying target features in the one type of similarity transformation images to determine that the check is a valid check, wherein the target feature is at least one of a seal or a signature; Characters in the two types of similarity transformation images are recognized to obtain text information in the image data.

6. The data processing method according to claim 1 or 4, characterized in that: Mapping the target information to a preset invoice template to obtain a target invoice corresponding to the check to be processed includes: According to the preset rules, the recognized text information is mapped with the fields in the target invoice template to obtain the field mapping result; Performing semantic recognition on the field mapping result, and determining the field attribution of the text information according to the semantic recognition result and the field mapping result; The text information is filled into the corresponding fields in the target invoice template according to the field attribution of the text information to obtain the target invoice corresponding to the to-be-processed check.

7. The data processing method according to claim 6, characterized in that: The performing semantic recognition on the field mapping result and determining the field attribution of the text information according to the semantic recognition result and the field mapping result includes: Checking the field mapping result to determine whether the text information is ambiguous; wherein the ambiguous situation is that the text information may be related to multiple different fields; Perform semantic analysis on ambiguous field mappings; Adjust ambiguous field mappings based on semantic analysis results to determine the field attribution of text information.

8. The data processing method according to claim 6, characterized in that: The target invoice template is determined from a plurality of preset invoice templates by at least one of the following methods: Performing template matching based on the position and type of different information areas in the image data, and determining a target invoice template corresponding to the check to be processed from a plurality of preset invoice templates; According to the user habits of the operating user, a target deposit slip template corresponding to the to-be-processed check is determined from a plurality of preset deposit slip templates.

9. A data processing device, characterized in that: The device comprises: An identification module, configured to perform anti-counterfeiting identification on the check to be processed based on three-dimensional imaging and / or multispectral imaging, and determine that the check to be processed is a genuine check; an extraction module, configured to extract text information from the check to be processed if the check to be processed is a genuine check; A mapping module is used to map the text information to a preset invoice template to obtain a target invoice corresponding to the check to be processed.

10. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

12. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images