Invoice processing method, processing device, electronic equipment and readable storage medium

Abstract

This disclosure provides a document processing method applicable to the fields of artificial intelligence and image recognition, as well as the financial field. The method includes: preprocessing m acquired target document images to obtain a first effective region image and a second effective region image corresponding to each target document image, where m is a positive integer greater than or equal to 1; extracting information from the first and second effective region images to obtain attribute information for each target document, wherein the attribute information includes identification information, which characterizes the target document type; and performing structured processing on the m target document images based on the document detail list information table and the identification information to obtain structured information for the m target document images. This disclosure also provides a document processing device, electronic device, readable storage medium, and program product.

Description

Technical Field

[0001] This disclosure relates to the fields of artificial intelligence and image recognition, and more specifically to a document processing method, processing device, electronic device, readable storage medium, and program product. Background Technology

[0002] Letters of credit are a commonly used settlement tool in international transactions, and the settlement process involves a large amount of document review and processing. Currently, staff at financial institution branches are responsible for receiving paper documents submitted by customers, verifying them against the document details list required by the letter of credit, and then digitizing the entire set of documents before submitting a document review application for processing.

[0003] In the process of realizing the present invention, the inventors discovered at least the following problems in the related technology: Currently, the paper documents submitted by customers need to be manually checked by the staff of financial institution branches, which is prone to verification errors. Moreover, when the branch staff upload electronic images of the documents as business attachments, they use a one-time upload method. Since there are many types of documents, and each type of document may have multiple copies, the upload lacks a hierarchical structure, which affects the review efficiency of the subsequent reviewers. Summary of the Invention

[0004] In view of the above problems, this disclosure provides a document processing method, processing apparatus, electronic device, readable storage medium, and program product.

[0005] According to one aspect of this disclosure, a document processing method is provided, comprising: preprocessing m acquired target document images to obtain a first effective area image and a second effective area image corresponding to each target document image, wherein m is a positive integer greater than or equal to 1; extracting information from the first effective area image and the second effective area image respectively to obtain attribute information of each target document, wherein the attribute information includes identification information, the identification information representing the target document type of the target document; and performing structured processing on the m target document images according to the document detail list information table and the identification information to obtain structured information for the m target document images.

[0006] According to an embodiment of this disclosure, preprocessing the acquired m target document images to obtain a first effective area image and a second effective area image corresponding to each target document image includes: numbering the m target document images to obtain m numbered target document images; calling an image processing tool to perform image segmentation on each numbered target document image according to preset screenshot parameters to obtain a first effective area image and a second effective area image corresponding to each numbered target document image.

[0007] According to embodiments of this disclosure, information extraction is performed on the first effective area image and the second effective area image to obtain attribute information for each target document, including: using a trained convolutional neural network to perform target image detection on the first effective area image to determine the identification information for the target image; using an OCR recognition model to extract information from the text content of the first effective area image and the text content of the second effective area image to obtain text information for the target document; and determining the attribute information for each target document based on the identification information and the text information.

[0008] According to embodiments of this disclosure, the attribute information of each target document further includes page number information.

[0009] According to an embodiment of this disclosure, the structured processing of m target document images based on a document detail list information table and identification information to obtain structured information for the m target document images includes: obtaining a document detail list information table, wherein the document detail list information table includes document types and the number of documents corresponding to each document type; determining identification information matching the document types in the document detail list information table; sorting the page number information of each target document corresponding to the identification information based on the identification information to determine the target document number corresponding to the identification information; and, if the target document number of the target document matches the document number in the document detail list information table, storing the target document images in a structured manner to obtain the structured information for the m target document images.

[0010] According to embodiments of this disclosure, obtaining a document detail list includes: obtaining the letter of credit number associated with the target document; and calling a business processing interface to obtain a document detail list that meets the requirements of the letter of credit terms based on the letter of credit number information.

[0011] According to embodiments of this disclosure, the method further includes: after obtaining the structured information of m target document images, sending a document review application request for the m target document images to the business processing system.

[0012] According to an embodiment of this disclosure, before preprocessing the acquired m target document images, the method further includes: in response to a request to acquire the m target document images, acquiring the m target document images sent by the electronic processing terminal.

[0013] Another aspect of this disclosure provides a document processing apparatus, comprising: a first processing module, configured to preprocess m acquired target document images to obtain a first effective area image and a second effective area image corresponding to each target document image, wherein m is a positive integer greater than or equal to 1; an extraction module, configured to extract information from the first effective area image and the second effective area image respectively to obtain attribute information for each target document, wherein the attribute information includes identification information, the identification information representing the target document type of the target document; and a second processing module, configured to perform structured processing on the m target document images according to a document detail list information table and the identification information to obtain structured information for the m target document images.

[0014] Another aspect of this disclosure provides an electronic device, including: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform the above-described document processing method.

[0015] Another aspect of this disclosure provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the aforementioned document processing method.

[0016] Another aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the above-described document processing method.

[0017] According to embodiments of this disclosure, by preprocessing the acquired m target document images, a first effective area image and a second effective area image corresponding to each target document image are obtained; information is extracted from the first and second effective area images respectively to obtain attribute information for each target document, wherein the attribute information includes identification information; based on the document detail list information table and the identification information, the m target document images are structured to obtain structured information for the m target document images. This solves the technical problems in related technologies, such as the cumbersome manual verification of document lists that meet the requirements of letter of credit terms, the tedious uploading of attachments when submitting document review applications, and the impact of disordered attachment information on the efficiency of subsequent document review by reviewers. It achieves intelligent analysis and structured organization of multiple target document images, realizing an ordered hierarchical structure for multiple target documents, reducing labor costs, and improving review efficiency and accuracy. Attached Figure Description

[0018] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0019] Figure 1This diagram illustrates an application scenario of the document processing method and apparatus according to embodiments of the present disclosure.

[0020] Figure 2 A flowchart illustrating a document processing method according to an embodiment of the present disclosure is shown schematically.

[0021] Figure 3 A flowchart illustrating a method for obtaining structured information for m target document images according to an embodiment of the present disclosure is shown.

[0022] Figure 4 A schematic block diagram of a document processing apparatus according to embodiments of the present disclosure is shown; and

[0023] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing a document processing method according to an embodiment of the present disclosure. Detailed Implementation

[0024] The embodiments of the present disclosure will now 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 disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0025] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0026] 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 are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0027] It should be noted that the document processing method and apparatus provided in this disclosure can be applied to the fields of artificial intelligence and image recognition, as well as the financial field, and can also be used in any technical field other than the financial field. The application fields of the business processing method and apparatus disclosed in this disclosure are not limited.

[0028] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0029] In realizing this disclosed concept, the inventor discovered that currently, financial institution branch staff receive documents from customers that meet the requirements of the letter of credit terms, manually verify these documents according to the letter of credit terms, and then digitize the verified documents, uploading the document images as attachments to the business processing system all at once for subsequent document review. However, this manual verification process by financial institution branch staff is prone to errors and is inefficient. When submitting document review applications to the business processing system and uploading document image files all at once, the lack of hierarchical organization among all document image files makes it difficult for subsequent document reviewers to quickly locate specific documents, further impacting review efficiency.

[0030] To address the issues arising from uploading document image attachments, financial institution branch staff can select each document individually for uploading. For example, they can select "Document List" and then upload the document image for the document list; select "Invoice" and then upload the document image for the invoice, and so on. However, this also increases the workload for financial institution branch staff, and requires them to upload various documents in page order, which affects their work efficiency.

[0031] Based on this, embodiments of this disclosure propose a document processing method, including: preprocessing m acquired target document images to obtain a first effective area image and a second effective area image corresponding to each target document image, where m is a positive integer greater than or equal to 1; extracting information from the first effective area image and the second effective area image respectively to obtain attribute information for each target document, wherein the attribute information includes identification information, and the identification information characterizes the target document type of the target document; and performing structured processing on the m target document images according to the document detail list information table and the identification information to obtain structured information for the m target document images.

[0032] It should be noted that in the technical solution disclosed herein, the acquisition, collection, storage, use, processing, transmission, provision, disclosure, and application of data all comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.

[0033] Figure 1The illustration shows an application scenario of the document processing method and processing apparatus according to embodiments of the present disclosure.

[0034] like Figure 1 As shown, application scenario 100 according to this embodiment may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as a medium for providing a communication link between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0035] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0036] Terminal devices 101, 102, and 103 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0037] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using terminal devices 101, 102, and 103 (for example only). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0038] It should be noted that the document processing method provided in this embodiment can generally be executed by server 105. Correspondingly, the document processing device provided in this embodiment can generally be located in server 105. The document processing method provided in this embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with terminal devices 101, 102, 103 and / or server 105. Correspondingly, the document processing device provided in this embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with terminal devices 101, 102, 103 and / or server 105.

[0039] For example, when a user uploads m target document images, terminal devices 101, 102, and 103 can send the m target document images to server 105, which then processes the m target document images to obtain structured information for them. Alternatively, a server or server cluster capable of communicating with terminal devices 101, 102, and 103 and / or server 105 can process the m target document images and ultimately obtain structured information for them.

[0040] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0041] Figure 2 A flowchart illustrating a document processing method according to an embodiment of the present disclosure is shown schematically.

[0042] like Figure 2 As shown, the method 200 includes operations S210 to S230.

[0043] In operation S210, the acquired m target document images are preprocessed to obtain the first effective area image and the second effective area image corresponding to each target document image, where m is a positive integer greater than or equal to 1.

[0044] In operation S220, information is extracted from the first effective area image and the second effective area image respectively to obtain the attribute information of each target document. The attribute information includes identification information, which represents the target document type of the target document.

[0045] In operation S230, based on the document detail list information table and identification information, the m target document images are processed in a structured manner to obtain structured information for the m target document images.

[0046] According to embodiments of this disclosure, image acquisition of the paper target document is required before obtaining the target document image. Image acquisition of the paper target document can be performed by scanning or photographing the target document using electronic processing devices such as scanners or cameras to obtain the target document image.

[0047] According to embodiments of this disclosure, operations such as grayscale conversion, binarization, noise reduction, tilt correction, and image smoothing can be performed on the acquired target document image to remove irrelevant content from the target document image and retain only the information of interest, thereby improving the reliability of processing the target document image.

[0048] According to embodiments of this disclosure, the target document can be any document that meets the requirements of the letter of credit terms, and can include various types, such as invoices, bills of lading, insurance policies, certificates of origin, quality inspection certificates, packing lists, etc. A letter of credit can be a conditional payment agreement and commitment made by a financial institution to the beneficiary.

[0049] According to embodiments of this disclosure, preprocessing the acquired m target document images may include using an image processing tool to process each document image to obtain a first effective area image and a second effective area image corresponding to each document image.

[0050] According to embodiments of this disclosure, the first effective area image and the second effective area image may include images that satisfy the preset area and preset threshold of the target document.

[0051] According to embodiments of this disclosure, a deep learning network model can be used to perform image recognition on the first effective area image and the second effective area image, so as to extract information from the text content of the first effective area image and the second effective area image and obtain the attribute information of each target document.

[0052] According to embodiments of this disclosure, the attribute information of the target documents in the first effective area image and the second effective area image are different. The first effective area image can represent the identification information, document number, document name, etc. of each target document, while the second effective area image can represent the page number information of each target document.

[0053] According to embodiments of this disclosure, the document detail list can represent a list of target documents that meet the requirements of the letter of credit terms. The document detail list can be collected and obtained by scanning or photographing the target documents simultaneously.

[0054] According to embodiments of this disclosure, the document detail list information table may include document types matching m target documents, and the number of documents corresponding to each document type. It should be noted that the number of documents may be the number of copies of that document type, and each document type may include multiple documents, with each document carrying a page number.

[0055] According to the embodiments of this disclosure, m target document images are arranged in the order of document types in the document details list information table, and the target documents of the same document type in the m target document images are structured to obtain structured information for the m target document images.

[0056] According to embodiments of this disclosure, the structured processing may include orderly hierarchical organization of m target document images, so that the disordered target document images form an ordered hierarchical structure.

[0057] According to embodiments of this disclosure, by preprocessing the acquired m target document images, a first effective area image and a second effective area image corresponding to each target document image are obtained; information is extracted from the first and second effective area images respectively to obtain attribute information for each target document, wherein the attribute information includes identification information; based on the document detail list information table and the identification information, the m target document images are structured to obtain structured information for the m target document images. This solves the technical problems in related technologies, such as the cumbersome manual verification of document lists that meet the requirements of letter of credit terms, the tedious uploading of attachments when submitting document review applications, and the impact of disordered attachment information on the efficiency of subsequent document review by reviewers. It achieves intelligent analysis and structured organization of multiple target document images, realizing an ordered hierarchical structure for multiple target documents, reducing labor costs, and improving review efficiency and accuracy.

[0058] According to embodiments of this disclosure, preprocessing the acquired m target document images to obtain a first effective area image and a second effective area image corresponding to each target document image may include:

[0059] Number the m target document images to obtain m numbered target document images; call the image processing tool, and according to the preset screenshot parameters, perform image segmentation on each numbered target document image to obtain the first effective area image and the second effective area image corresponding to each numbered target document image.

[0060] According to the embodiments of this disclosure, m target document images can be numbered from 1 to m. It should be noted that if the document detail list image corresponding to the document detail list information table and the target document image are obtained at the same time, the document detail list image is identified by the deep learning network model and numbered as 0.

[0061] According to embodiments of this disclosure, the image processing tool may include OpenCV. By calling the OpenCV image processing tool, image segmentation is performed on target document images with a number greater than or equal to 1 based on pre-set screenshot parameters. For each target document image, it can be segmented into a first effective area image and a second effective area image.

[0062] According to embodiments of this disclosure, in practice, information such as the title, document number, and document identifier of a document image is typically set in a specific area of ​​the upper half of the image, while the page number information is set in a specific area of ​​the lower half. The target document image is then segmented according to pre-set area screenshot parameters. For example, the screenshot parameters can be set to 20-25% of the upper half and 10-15% of the lower half. It should be noted that the screenshot parameters can be flexibly adjusted according to actual image segmentation needs, and are not limited here.

[0063] According to embodiments of this disclosure, the first valid area image can be denoted as number _A, and the second valid area image can be denoted as number _B. For a target document image with number 1, the first valid area image can be denoted as 1_A, the second valid area image of the target document image with number 1 can be denoted as 1_B, and so on. For a target document image with number m, the first valid area image can be denoted as m_A, and the second valid area image of the target document image with number m can be denoted as m_B.

[0064] According to embodiments of this disclosure, information is extracted from the first effective area image and the second effective area image respectively to obtain attribute information for each target document, including:

[0065] The trained convolutional neural network is used to detect target images in the first effective area image to determine the identification information of the target image; the OCR recognition model is used to extract information from the text content of the first effective area image and the text content of the second effective area image to obtain the text information of the target document; the attribute information of each target document is determined based on the identification information and the text information.

[0066] According to embodiments of this disclosure, the first effective region image is input into a trained convolutional neural network model to extract the identification information of the target image. The identification information of the target image can characterize the document type of the target document image. For example, the identification information corresponding to an image of document type invoice can be A007, and the identification information corresponding to an image of document type bill of lading can be A008, etc.

[0067] According to embodiments of this disclosure, the model for extracting the identifier information of the target image in the target document image may include convolutional neural network models such as YOLO (You Only LookOnce), R-CNN (Region-CNN), and Fast R-CNN (FastRegion-CNN). In practical applications of the embodiments of this disclosure, the specific algorithm used may be adjusted according to actual circumstances, and the embodiments of this disclosure do not limit this.

[0068] According to embodiments of this disclosure, an OCR recognition model can be used to recognize and extract text information from a first valid area image and a second valid area image. The text information for the first valid area image may include the document name, document number, and document type recognition result of the target document image; the text information for the second valid area image may include the page number of the target document image. For example, Table 1 schematically shows the recognition result for target document image number 1. (See Table 1.)

[0069] Table 1

[0070]

[0071] According to embodiments of this disclosure, a convolutional neural network and an OCR recognition model are used to iteratively process m target document images, outputting the recognition results of attribute information for each target document image. For example, Table 2 schematically shows the recognition results for each target document image. As shown in Table 2.

[0072] Table 2

[0073]

[0074]

[0075] As shown in Table 3, “Y” indicates that the document type of the corresponding target document was successfully identified; “N” indicates that the document type of the corresponding target document was not identified; and “ / ” indicates that the document number of the target document was not identified.

[0076] According to the embodiments of this disclosure, when the target information cannot be identified in the first effective area image and the second effective area image, the target information can be extracted from the entire target document image to complete the identification of the document type and page number information of the target document image.

[0077] According to embodiments of this disclosure, for target document images in Table 2 whose document types are identified as failed, manual confirmation can be performed by personnel at financial institution branches.

[0078] Figure 3 A flowchart illustrating a method for obtaining structured information for m target document images according to an embodiment of the present disclosure is shown.

[0079] like Figure 3 As shown, the method 300 may include operations S310 to S340.

[0080] In operation S310, the document details list information table is obtained. The document details list information table includes the document type and the number of documents corresponding to the document type.

[0081] In operation S320, based on the document type in the document details list information table, determine the identification information that matches the document type in the document details list information table.

[0082] In operation S330, based on the identification information, the page number information of each target document corresponding to the identification information is sorted to determine the number of target documents corresponding to the identification information.

[0083] In operation S340, if the number of target documents and the number of documents in the document detail list information table are consistent, the target document images are stored in a structured manner to obtain structured information for m target document images.

[0084] According to embodiments of this disclosure, obtaining a document detail list may include: obtaining the letter of credit number associated with the target document; and calling a business processing interface to obtain the document detail list that meets the requirements of the letter of credit terms based on the letter of credit number information.

[0085] According to embodiments of this disclosure, in response to an input request from a financial institution branch staff member, the letter of credit number information associated with the target document, entered by the branch staff member, is obtained. The letter of credit number information is unique; each letter of credit corresponds to a single number.

[0086] According to embodiments of this disclosure, a business processing interface can be invoked to retrieve a list of document details corresponding to the letter of credit number information that meets the requirements of the letter of credit terms from the business processing system. For example, Table 3 schematically illustrates a list of document details according to an embodiment of this disclosure. (See Table 3.)

[0087] Table 3

[0088] Document Type Number of copies bill 1 Bill of lading 2 insurance policy 1 Certificate of Origin 1 Quality inspection certificate 1 Packing list 1 ......

[0089] As shown in Table 3, the document details list can include the document type and the corresponding quantity, i.e., the number of copies of that document type. Table 3 shows document types and quantities as examples; the specific requirements may vary depending on the letter of credit terms for the specific business transaction.

[0090] According to embodiments of this disclosure, the identification information matching the document type in the document detail list information table can be determined sequentially in the recognition result for each target document image, according to the order of document types. For example, firstly, based on the document type "invoice" in the document detail list information table, the identification information corresponding to the document type "invoice" in the recognition result of each target document image is determined, as shown in "A007" in Table 2.

[0091] According to embodiments of this disclosure, page number information for each target document image can be determined based on the identification information corresponding to the determined document type. The page number information of the target document images corresponding to the identification information can be sorted according to a pre-set sorting rule. The pre-set sorting rule can be to sort the page number information in ascending order, and the sorted target document images corresponding to the identification information can be considered as a single document of that document type.

[0092] According to the embodiments of this disclosure, for example, referring to Table 2, when the identification information corresponding to the document type "invoice" is "A007", the page number information of the target document image corresponding to "A007" is sorted in ascending order, i.e. "1, 2, 3, ..., 8", and the target document images of pages 1-8 corresponding to the identification information "A007" can be regarded as an invoice.

[0093] According to embodiments of this disclosure, the method for determining the target number of other document types is similar to the method for determining the "invoice" document type, and will not be described again here.

[0094] According to the embodiments of this disclosure, when determining whether the quantity of the target document is consistent with the quantity in the document detail list information table, a pre-set hit verification rule can be used for verification, and corresponding prompt information can be output, such as shown in Table 4.

[0095] Table 4

[0096]

[0097] According to embodiments of this disclosure, when the number of target documents in the target document matches the number of documents in the document detail list information table—that is, when the target document image verification is completely successful—the target document image is stored in a structured manner. The structured storage may include storage in a tree structure, thereby displaying the structured information of each document of each document type. For example, it may be displayed in the following output format.

[0098] Letter of Credit No.: LC90010220022123

[0099] 1. Document List

[0100] 2 Letter of Credit Text

[0101] 2.1 Letter of Credit Text Page 1

[0102] 2.2 Letter of Credit Text Page 2 ...

[0104] 2.n Letter of Credit Text Page n

[0105] 3 Invoices

[0106] 3.1 Invoice Page 1

[0107] 3.2 Invoice Page 2

[0108] 4. Bills of Lading (First Set)

[0109] 4.1 Bill of Lading (First Set) Page 1

[0110] 4.2 Bill of Lading (First Set) Page 2 ...

[0112] 4.n Bill of Lading (First Set) Page n

[0113] 5. Bill of Lading (Second Set)

[0114] 5.1 Bill of Lading (Second Set) Page 1

[0115] 5.2 Bill of Lading (Second Set) Page 2 ...

[0117] 5.n Bill of Lading (Second Set) Page n

[0118] 6 insurance policies

[0119] 6.1 Policy Page 1

[0120] 6.2 Policy Page 2 ...

[0122] 6.n policy page n

[0123] 7 Certificates of Origin

[0124] 7.1 Certificate of Origin Page 1

[0125] 7.2 Certificate of Origin Page 2 ...

[0127] 7. Certificate of Origin Page

[0128] 8 Quality Inspection Certificates

[0129] 8.1 Quality Inspection Certificate Page 1

[0130] 8.2 Quality Inspection Certificate Page 2 ...

[0132] 8.n Quality Inspection Certificate Page

[0133] 9 Packing List

[0134] 9.1 Packing List 1

[0135] 9.2 Packing List Page 2

[0136] 9. Packing list page

[0137] ...

[0138] According to the embodiments of this disclosure, after organizing m target document images into a tree structure, when subsequent document reviewers access the images, they can quickly locate the corresponding document page and view the document information by clicking on the page number information of the document type corresponding to the structure tree, thereby improving the work efficiency of subsequent document reviewers.

[0139] According to an embodiment of this disclosure, after obtaining the structured information of m target document images, a document review application request for the m target document images is sent to the business processing system.

[0140] According to the embodiments of this disclosure, if the m target document images are completely consistent, a document review application request for the m target document images is sent to the business processing system under the corresponding letter of credit number information.

[0141] According to embodiments of this disclosure, before preprocessing the acquired m target document images, the method further includes: in response to a request to acquire the m target document images, acquiring the m target document images sent by the electronic processing terminal.

[0142] According to embodiments of this disclosure, the acquisition request can be a request to acquire a target document image, and the electronic processing terminal can be a system for acquiring the target document image.

[0143] Based on the above document processing method, this disclosure also provides a document processing apparatus. The following will be combined with... Figure 4 The device is described in detail.

[0144] Figure 4 A schematic block diagram of a document processing apparatus according to an embodiment of the present disclosure is shown.

[0145] like Figure 4 As shown, the document processing device 400 of this embodiment may include: a first processing module 410, an extraction module 420, and a second processing module 430.

[0146] The first processing module 410 is used to preprocess the acquired m target document images to obtain a first effective area image and a second effective area image corresponding to each target document image, where m is a positive integer greater than or equal to 1. In one embodiment, the first processing module 410 can be used to perform the operation S210 described above, which will not be repeated here.

[0147] The extraction module 420 is used to extract information from the first effective area image and the second effective area image respectively to obtain the attribute information of each target document. The attribute information includes identification information, which represents the target document type. In one embodiment, the extraction module 420 can be used to perform the operation S220 described above, which will not be repeated here.

[0148] The second processing module 430 is used to perform structured processing on m target document images based on the document detail list information table and identification information, to obtain structured information for the m target document images. In one embodiment, the second processing module 430 can be used to execute the operation S230 described above, which will not be repeated here.

[0149] According to embodiments of this disclosure, by preprocessing the acquired m target document images, a first effective area image and a second effective area image corresponding to each target document image are obtained; information is extracted from the first and second effective area images respectively to obtain attribute information for each target document, wherein the attribute information includes identification information; based on the document detail list information table and the identification information, the m target document images are structured to obtain structured information for the m target document images. This solves the technical problems in related technologies, such as the cumbersome manual verification of document lists that meet the requirements of letter of credit terms, the tedious uploading of attachments when submitting document review applications, and the impact of disordered attachment information on the efficiency of subsequent document review by reviewers. It achieves intelligent analysis and structured organization of multiple target document images, realizing an ordered hierarchical structure for multiple target documents, reducing labor costs, and improving review efficiency and accuracy.

[0150] According to embodiments of this disclosure, the first processing module 410 may include: a numbering submodule and a calling submodule.

[0151] The numbering submodule is used to number m target document images to obtain m numbered target document images.

[0152] The submodule is called to invoke the image processing tool, which performs image segmentation on each numbered target document image according to preset screenshot parameters, to obtain the first effective area image and the second effective area image corresponding to each numbered target document image.

[0153] According to embodiments of this disclosure, the extraction module 420 may include: a detection submodule, an extraction submodule, and a determination submodule.

[0154] The detection submodule is used to perform target image detection on the first effective region image using a trained convolutional neural network and determine the identification information for the target image.

[0155] The extraction submodule is used to extract information from the text content of the first effective area image and the text content of the second effective area image using an OCR recognition model, so as to obtain the text information for the target document.

[0156] The first determination submodule is used to determine the attribute information of each target document based on the identification information and text information.

[0157] According to embodiments of this disclosure, the attribute information of each target document also includes page number information.

[0158] According to embodiments of this disclosure, the second processing module 430 may include: an acquisition submodule, a second determination submodule, a sorting submodule, and a storage submodule.

[0159] The acquisition submodule is used to acquire the document detail list information table, which includes the document type and the number of documents corresponding to the document type;

[0160] The second determination submodule is used to determine the identification information that matches the document type in the document details list information table based on the document type in the document details list information table.

[0161] The sorting submodule is used to sort the page number information of each target document corresponding to the identification information according to the identification information, and determine the number of target documents corresponding to the identification information.

[0162] The storage submodule is used to perform structured storage of the target document images when the number of target documents and the number of documents in the document detail list information table are consistent, so as to obtain structured information for m target document images.

[0163] According to embodiments of this disclosure, the acquisition submodule may include: an acquisition unit and a calling unit.

[0164] The acquisition unit is used to acquire the letter of credit number information associated with the target document.

[0165] The calling unit is used to call the business processing interface and obtain a list of document details that meet the requirements of the letter of credit terms based on the letter of credit number information.

[0166] According to embodiments of this disclosure, the document processing apparatus 400 may further include a sending module.

[0167] The sending module is used to send a document review application request for the m target document images to the business processing system after obtaining the structured information of the m target document images.

[0168] According to embodiments of this disclosure, the document processing apparatus 400 may further include a response module.

[0169] The response module is used to respond to a request to acquire m target document images and acquire the m target document images sent by the electronic processing terminal.

[0170] According to embodiments of this disclosure, any plurality of modules among the first processing module 410, extraction module 420, and second processing module 430 may be combined into one module, or any one of these modules may be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules may be combined with at least some of the functions of other modules and implemented in one module.

[0171] According to embodiments of this disclosure, at least one of the first processing module 410, the extraction module 420, and the second processing module 430 can be at least partially implemented as hardware circuitry, 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-package, an application-specific integrated circuit (ASIC), or any other reasonable method of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three methods. Alternatively, at least one of the first processing module 410, the extraction module 420, and the second processing module 430 can be at least partially implemented as a computer program module, which, when run, can perform corresponding functions.

[0172] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing a document processing method according to an embodiment of the present disclosure.

[0173] like Figure 5 As shown, an electronic device 500 according to an embodiment of the present disclosure includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage portion 508 into a random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0174] RAM 503 stores various programs and data required for the operation of electronic device 500. Processor 501, ROM 502, and RAM 503 are interconnected via bus 504. Processor 501 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 502 and / or RAM 503. It should be noted that the programs may also be stored in one or more memories other than ROM 502 and RAM 503. Processor 501 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0175] According to embodiments of this disclosure, the electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to a bus 504. The electronic device 500 may also include one or more of the following components connected to the I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 510 as needed so that computer programs read from it can be installed into the storage section 508 as needed.

[0176] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0177] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the 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 embodiments of this disclosure, the computer-readable storage medium may include ROM 502 and / or RAM 503 and / or one or more memories other than ROM 502 and RAM 503 described above.

[0178] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement the item recommendation method provided in the embodiments of this disclosure.

[0179] When the computer program is executed by the processor 501, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0180] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 509, and / or installed from a removable medium 511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0181] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by processor 501, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0182] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational 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 execute entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0183] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0184] Those skilled in the art will understand that the features described in the various embodiments of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments of this disclosure can be combined and / or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0185] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A bill processing method, comprising: preprocessing m obtained target bill images to obtain a first effective area image and a second effective area image corresponding to each of the target bill images, wherein m is a positive integer greater than or equal to 1, the first effective area image represents identification information, a bill number, and a bill name of each of the target bill images, and the second effective area image represents page number information of each of the target bill images; respectively extracting information from the first effective area image and the second effective area image to obtain attribute information of each target bill image, wherein the attribute information includes the identification information and the page number information of the target bill image, and the identification information represents a target bill type of the target bill image; performing structured processing on the m target bill images according to a bill list information table and the identification information to obtain structured information for the m target bill images, including: obtaining the bill list information table, wherein the bill list information table includes a bill type and a bill quantity corresponding to the bill type; determining the identification information matching the bill type in the bill list information table according to the bill type in the bill list information table; determining the page number information of each of the target bill images according to the determined identification information matching the bill type, sorting the page number information of the target bill image corresponding to the identification information, and determining a target bill corresponding to the identification information as a bill corresponding to the bill type after sorting the target bill image corresponding to the identification information, and determining a target bill quantity of the target bill corresponding to the identification information; in a case where the target bill quantity and the bill quantity in the bill list information table are consistent, performing structured storage on the target bill image to obtain the structured information for the m target bill images.

2. The method of claim 1, wherein, The preprocessing of the m obtained target bill images to obtain the first effective area image and the second effective area image corresponding to each of the target bill images includes: numbering the m target bill images to obtain m numbered target bill images; calling an image processing tool to perform image cutting on each of the numbered target bill images according to preset screenshot parameters to obtain the first effective area image and the second effective area image corresponding to each of the numbered target bill images.

3. The method of claim 1, wherein, The respective extraction of information from the first effective area image and the second effective area image to obtain the attribute information of each target bill includes: performing target image detection on the first effective area image by using a trained convolutional neural network to determine the identification information of the target image; respectively extracting information from text content of the first effective area image and text content of the second effective area image by using an OCR recognition model to obtain text information of the target bill; determining the attribute information of each of the target bills according to the identification information and the text information.

4. The method of claim 1, wherein, The obtaining of the bill detail list information table comprises: obtaining the credit number information associated with the target bill; calling a business processing interface to obtain the bill detail list information table meeting the credit clause requirements according to the credit number information.

5. The method of claim 1, further comprising: after obtaining the structured information of the m target bill images, sending a bill auditing business application request for the m target bill images to a business processing system.

6. The method of claim 1, before preprocessing the obtained m target bill images, further comprising: in response to the request for the m target bill images, obtaining the m target bill images sent by an electronic processing terminal.

7. A bill processing apparatus, comprising: a first processing module configured to preprocess the obtained m target bill images to obtain a first effective area image and a second effective area image corresponding to each of the target bill images, wherein m is a positive integer greater than or equal to 1, the first effective area image represents the identification information, bill number, and bill name of each of the target bill images, and the second effective area image represents the page number information of each of the target bill images; an extraction module configured to extract information from the first effective area image and the second effective area image respectively to obtain attribute information of each of the target bill images, wherein the attribute information comprises the identification information and the page number information of the target bill image, and the identification information represents the target bill type of the target bill image; a second processing module configured to perform structured processing on the m target bill images according to a bill detail list information table and the identification information to obtain structured information of the m target bill images, comprising: obtaining the bill detail list information table, wherein the bill detail list information table comprises a bill type and a bill quantity corresponding to the bill type; determining the identification information matching the bill type in the bill detail list information table according to the bill type in the bill detail list information table; determining the page number information of each of the target bill images according to the determined identification information matching the bill type, sorting the page number information of the target bill image corresponding to the identification information, and determining a target bill corresponding to the identification information as a bill corresponding to the bill type according to the sorted identification information, to determine a target bill quantity of the target bill corresponding to the identification information; in a case where the target bill quantity and the bill quantity in the bill detail list information table are consistent, performing structured storage on the target bill images to obtain the structured information of the m target bill images.

8. An electronic device, comprising: one or more processors; a storage device configured to store one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method of any one of claims 1-6.

10. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-6.

Images