WO2024012209A1

WO2024012209A1 - Image recognition-based service processing method and apparatus, and storage medium

Info

Publication number: WO2024012209A1
Application number: PCT/CN2023/103596
Authority: WO
Inventors: 彭波
Original assignee: 深圳前海环融联易信息科技服务有限公司
Priority date: 2022-07-13
Filing date: 2023-06-29
Publication date: 2024-01-18
Also published as: CN115019325A

Abstract

The present application relates to the technical field of computers. Disclosed are an image recognition-based service processing method and apparatus, a storage medium, and a computer device. The method comprises: acquiring at least one image to be verified, and performing character recognition on said at least one image to obtain character information corresponding to each image to be verified, each image to be verified comprising at least one invoice to be verified; determining from the character information target query information corresponding to each invoice to be verified, and on the basis of the target query information, determining whether the invoice to be verified has passed verification; and when any invoice to be verified has passed verification, on the basis of the at least one image to be verified, determining a target archived image corresponding to said invoice to be verified, and generating archived information of a service on the basis of the target archived image. The present application can simplify a verification process of an invoice to be verified, thus greatly improving verification and archiving efficiency, and improving experience of a user.

Description

A business processing method, device and storage medium based on image recognition

This application requests the priority of the Chinese patent application submitted to the China Patent Office on July 13, 2022, with the application number 202210821064.2 and the application title "A business processing method, device and storage medium based on image recognition", and its entire content Incorporated into the application by reference.

Technical field

The present application relates to the field of computer technology, and in particular, to a business processing method and device based on image recognition, storage media, and computer equipment.

Background technique

Currently, many businesses require verification of documents during the processing process. For example, when using documents such as invoices for financing, the invoice needs to be verified. Only after verification shows that the invoice has not been used for financing can the invoice be used. The invoice is used to handle this financing business. However, in the existing technology, when verifying invoices before financing, users usually need to log in to invoice verification systems such as the Zhongdeng system offline and enter invoice-related information multiple times for verification, which makes the verification process cumbersome. The inventor realized that after the invoice passed the verification, the relevant information of the verified invoice was manually uploaded to the financing system to handle the financing business, and the user needed to find the invoice used to handle the financing business from many invoices and upload it for archiving. The whole process was very complicated. The workload is heavy, the efficiency is low, and the user experience is poor.

Contents of the invention

In view of this, this application provides a business processing method and device, storage medium, and computer equipment based on image recognition, which can simplify the verification process of documents to be verified, greatly improve the efficiency of verification and archiving, and help improve the user experience. feel.

According to one aspect of this application, a business processing method based on image recognition is provided, including:

Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each of the images to be verified, wherein each of the images to be verified includes at least one document to be verified;

Determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

After any of the documents to be verified passes the verification, a target archive image corresponding to any of the documents to be verified is determined based on the at least one image to be verified, and the business is generated based on the target archive image archived information.

According to another aspect of the present application, a service processing device based on image recognition is provided, including:

A text recognition module, configured to obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each image to be verified, wherein each image to be verified includes at least one Documents to be verified;

A verification module, configured to determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

An archiving module, configured to determine a target archive image corresponding to any of the documents to be verified based on the at least one image to be verified after the document to be verified passes the verification, and based on the target archive image , generate archive information of the business.

According to another aspect of the present application, a storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the following steps are implemented:

Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each image to be verified, wherein each image to be verified includes at least one document to be verified;

Determine the target query information corresponding to each document to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

When any document to be verified passes the verification, a target archive image corresponding to any document to be verified is determined based on at least one image to be verified, and based on the target archive image, archive information of the business is generated.

According to yet another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor. When the processor executes the program, it implements the following steps:

With the above technical solution, this application provides a business processing method and device, storage medium, and computer equipment based on image recognition. First, one or more images to be verified can be obtained, and each image to be verified can contain one or Multiple documents to be verified. After obtaining one or more images to be verified, text recognition can be performed on each image to be verified, and the text information corresponding to each image to be verified can be obtained. After obtaining the text information corresponding to each image to be verified, when the image to be verified contains a document to be verified, the target query information corresponding to the document to be verified can be determined directly from the text information corresponding to the image to be verified; when the document to be verified is to be When the image contains multiple documents to be verified, the target query information corresponding to each document to be verified can be determined from the text information corresponding to the image to be verified. After the target query information is determined, each document to be verified can be verified based on the target query information corresponding to the document to be verified, and it is determined whether the document to be verified can pass the verification. If there is a document to be verified that has passed the verification, the document to be verified that has passed the verification can be used to handle related business in the future. During the business processing process, in order to complete business-related information, documents to be verified can also be archived. Specifically, the target archive image corresponding to the verified document to be verified can be found from the obtained image to be verified, and then the archive information of the business can be generated based on the target archive image. The embodiments of this application can simplify the verification process of documents to be verified, greatly improve the efficiency of verification and archiving, and help improve the user experience.

The above description is only an overview of the technical solutions of the present application. In order to have a clearer understanding of the technical means of the present application, they can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and understandable. , the specific implementation methods of the present application are specifically listed below.

Description of drawings

The drawings described here are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the attached picture:

Figure 1 shows a schematic flow chart of a business processing method based on image recognition provided by an embodiment of the present application;

Figure 2 shows a schematic flow chart of another image recognition-based business processing method provided by an embodiment of the present application;

Figure 3 shows a schematic structural diagram of a service processing device based on image recognition provided by an embodiment of the present application.

Detailed ways

The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, as long as there is no conflict, the embodiments and features in the embodiments of this application can be combined with each other.

In this embodiment, a business processing method based on image recognition is provided. As shown in Figure 1, the method includes:

Step 101: Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each of the images to be verified, wherein each of the images to be verified includes at least one document to be verified. .

The business processing method based on image recognition provided by the embodiment of this application can be applied to invoice financing business scenarios. First, one or more images to be verified can be obtained, and each image to be verified can contain one or more documents to be verified. Here, the document to be verified can be an invoice, and the image to be verified can be a picture or pdf of an invoice. When multiple invoices are posted in one pdf file, or multiple invoices are in the same picture, the image to be verified can be It is a picture or pdf composed of multiple invoices. When the image to be verified is a picture, the image to be verified can be obtained by photographing a paper invoice with a camera; when the image to be verified is a PDF, the image to be verified can be obtained by scanning with a scanner, or it can be obtained directly from the electronic invoice. Downloadable from the website. After obtaining one or more images to be verified, text recognition can be performed on each image to be verified, and the text information corresponding to each image to be verified can be obtained. Specifically, OCR technology can be used to achieve this. OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition). It converts the text of various bills, newspapers, books, manuscripts and other printed matter into image information through optical input methods such as scanning, and then uses text recognition technology to convert the image information. Enter technology for computers that can be used.

Step 102: Determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information.

In this embodiment, after obtaining the text information corresponding to each image to be verified, when the image to be verified contains a document to be verified, the target corresponding to the document to be verified can be determined directly from the text information corresponding to the image to be verified. Query information; when the image to be verified contains multiple documents to be verified, the target query information corresponding to each document to be verified can be determined from the text information corresponding to the image to be verified. After the target query information is determined, each document to be verified can be verified based on the target query information corresponding to the document to be verified, and it is determined whether the document to be verified can pass the verification. Here, when the document to be verified is an invoice, the target query information can be the invoice name, company name, trade The contract name, invoice number, invoice amount, invoice date, tax-included amount, tax-exclusive amount, special invoice, general invoice, seller's name, buyer's name, etc. can be selected according to the actual situation. When exact matching is desired, the target query information can be determined to be multiple of the above. This application can directly obtain the target query information corresponding to each document to be verified based on the obtained image to be verified, and automatically input the target query information into the document verification system. There is no need for users to log in to the document verification system offline and manually enter the target multiple times. Query information for verification, greatly simplifying the verification process.

Step 103: After any of the documents to be verified passes the verification, determine the target archive image corresponding to any of the documents to be verified based on the at least one image to be verified, and generate based on the target archive image. Archived information for said business.

In this embodiment, if there is a document to be verified that has passed the verification, the document to be verified that has passed the verification can be used to handle related business later. During the business processing process, in order to complete business-related information, documents to be verified can also be archived. Specifically, the target archive image corresponding to the verified document to be verified can be found from the obtained image to be verified, and then the archive information of the business can be generated based on the target archive image. When the image to be verified includes multiple documents to be verified, and not all documents to be verified have passed verification, that is, not all documents to be verified can be used for business processing, if the document to be verified is manually found from multiple images to be verified. document, and extracting the verified document to be verified from the image to be verified, and then uploading it for archiving, which requires a lot of time, low efficiency, and poor user experience. The embodiment of the present application can directly and automatically extract the document from multiple documents. The document to be verified for business processing is determined in the image to be verified, which is simple, convenient and efficient, and is conducive to improving the user experience. For example, in an invoice financing scenario, users are usually required to find documents to be verified (i.e. invoices) that have passed verification from many images to be verified, and then upload them manually. However, this series of processes can be automatically implemented, which greatly improves archiving efficiency. and accuracy.

By applying the technical solution of this embodiment, first, one or more images to be verified can be obtained, and each image to be verified can contain one or more documents to be verified. After obtaining one or more images to be verified, text recognition can be performed on each image to be verified, and the text information corresponding to each image to be verified can be obtained. After obtaining the text information corresponding to each image to be verified, when the image to be verified contains a document to be verified, the target query information corresponding to the document to be verified can be determined directly from the text information corresponding to the image to be verified; when the document to be verified is to be When the image contains multiple documents to be verified, the target query information corresponding to each document to be verified can be determined from the text information corresponding to the image to be verified. After the target query information is determined, each document to be verified can be verified based on the target query information corresponding to the document to be verified, and it is determined whether the document to be verified can pass the verification. If there is a document to be verified that has passed the verification, the document to be verified that has passed the verification can be used to handle related business in the future. During the business processing process, in order to complete business-related information, documents to be verified can also be archived. Specifically, the target archive image corresponding to the verified document to be verified can be found from the obtained image to be verified, and then the archive information of the business can be generated based on the target archive image. The embodiments of this application can simplify the verification process of documents to be verified, greatly improve the efficiency of verification and archiving, and help improve the user experience.

Further, as a refinement and expansion of the specific implementation of the above embodiment, in order to completely explain the specific implementation process of this embodiment, another business processing method based on image recognition is provided. As shown in Figure 2, the method includes:

Step 201: Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each image to be verified, wherein each image to be verified includes at least one document to be verified. .

In this embodiment, first, one or more images to be verified can be obtained, and each image to be verified can contain one or more documents to be verified. Here, the document to be verified can be an invoice, and the image to be verified can be an invoice. A picture or pdf composed of invoices. When multiple invoices are posted in one pdf file, or multiple invoices are in the same picture, the image to be verified can be a picture or pdf composed of multiple invoices. When the image to be verified is a picture, the image to be verified can be obtained by photographing a paper invoice with a camera; when the image to be verified is a PDF, the image to be verified can be obtained by scanning with a scanner, or it can be obtained directly from the electronic invoice. Downloadable from the website. After obtaining one or more images to be verified, text recognition can be performed on each image to be verified, and the text information corresponding to each image to be verified can be obtained.

Step 202: Based on the text information, determine the number of documents to be verified contained in the image to be verified.

In the embodiment of the present application, optionally, step 202 includes: identifying the number of occurrences of the first word combination from the text information, and using the number of occurrences as the individual number of the document to be verified contained in the image to be verified. number, wherein the first character combination is a character combination that exists in each of the documents to be verified and appears only once.

In this embodiment, after obtaining the text information corresponding to each image to be verified, if the image to be verified includes multiple documents to be verified, the text information may include multiple repeated information. Specifically, if an image to be verified is If the corresponding text information contains a first text combination, then it can be determined that the image to be verified contains a document to be verified; if the text information corresponding to a certain image to be verified contains multiple first text combinations, then it can be determined that the image to be verified is Contains the number of occurrences of the first word combination in documents to be verified. Here, the first text combination may be a text combination that appears in each document to be verified and appears only once. For example, when the document to be verified is an invoice, the first text combination may be "Special Value Added Tax Invoice", "Password Area" " and other text combinations.

Step 203: When a document to be verified is included, a unique identifier is determined from the text information, the image to be verified is marked with the unique identifier, and the marked image to be verified is used as the target verification image.

In this embodiment, if the image to be verified contains only one document to be verified, the unique identifier corresponding to the document to be verified can be found from the text information corresponding to the image to be verified. After that, the unique identifier can be used to verify the document to be verified. The image is marked, and the image to be verified after the marking process can be used as the target verification image. The unique identifier may be an identifier that corresponds one-to-one with the document to be verified. When the document to be verified is an invoice, the unique identifier may be an invoice number, etc.

Step 204: When multiple documents to be verified are included, the image to be verified is cropped to obtain a sub-image to be verified corresponding to each document to be verified, and the sub-image to be verified is determined from the text information. The sub-images to be verified are marked with unique identifiers corresponding to the sub-images to be verified, and each marked sub-image to be verified is used as the target verification image.

In this embodiment, if the image to be verified contains multiple documents to be verified, the image to be verified can be cropped to obtain multiple sub-images to be verified, where each sub-image to be verified corresponds to a document to be verified. In addition, the unique identifier corresponding to each sub-image to be verified can also be determined from the text information corresponding to the image to be verified. Afterwards, the obtained unique identifier can be used to mark the corresponding sub-image to be verified, and the marked sub-image to be verified can be used as the target verification image.

Step 205: Determine target query information corresponding to each document to be verified from the text information.

In the embodiment of the present application, optionally, when the image to be verified contains multiple documents to be verified, step 205 includes: dividing the text information into sub-sections corresponding to each of the target verification images. text information, and extract the target query information corresponding to the document to be verified from each sub-text information.

In this embodiment, when the image to be verified contains only one document to be verified, the target query information of the document to be verified can be determined directly from the text information corresponding to the image to be verified. When the image to be verified contains multiple documents to be verified, the text information can be divided into sub-text information corresponding to each target verification image, and the target query information corresponding to the document to be verified can be further extracted from each sub-text information. . For example, if the image to be verified contains 4 documents to be verified, the text information corresponding to the image to be verified can be divided into 4 sub-text information, and then the target query corresponding to each document to be verified can be extracted from each sub-text information. information.

Step 206: Call the preset document verification interface, input the target query information into the preset document verification interface, and receive the document status information returned by the preset document verification interface, where the document status information includes the Registered status and unregistered status; when the document status information is in the registered status, it is determined that the document to be verified has not passed verification; when the document status information is in the unregistered state, it is determined that the document to be verified has passed verification.

In this embodiment, after determining the target query information corresponding to each document to be verified, a preset document verification interface can be called. By calling the preset document verification interface, the document to be verified can be verified. After calling the preset document verification interface, you can input the target query information corresponding to each document to be verified into the preset document verification interface, and you can also receive the document status information returned by the preset document verification interface. For example, when the document to be verified is an invoice, the default document verification interface may be the Zhonglog query system call interface. This application calls the Zhongdeng query system that needs to be used for invoice financing business directly through a third-party interface call, which can avoid users' manual cross-system query. After querying, the verification information is then manually input into the financing system to handle the financing business, which can greatly improve Verification efficiency. The document status information returned after verification can include registered status and unregistered status. If the document status information is registered, it means that the document to be verified has been used to handle the business and can no longer be used to handle the business. At this time, it can be determined that the document to be verified has not passed the verification; if the document status information is unregistered , then it means that the document to be verified has not been used to handle this business, but can be used to handle this business. At this time, it can be determined that the document to be verified has passed the verification. For example, in the invoice financing scenario, the document to be verified at this time can be an invoice, and the business can be a financing business. If the document status information is registered, it means that the invoice has been used for financing business and cannot be used for financing again. , if the document status information is unregistered, it means that the invoice has not been used for financing business, and the invoice can be used for financing.

Step 207: After any of the documents to be verified passes the verification, determine the document corresponding to any of the documents to be verified from the target verification image corresponding to the at least one image to be verified based on the unique identifier. The target verification image serves as the target archive image, and based on the target archive image, the archive information of the business is generated.

In this embodiment, if there is a document to be verified that passes the verification, the document to be verified can be used to handle the business. After the business is handled, the target archive image corresponding to the document to be verified can also be used for archiving. Here, the target archive image is the image corresponding to the document to be verified that needs to be archived. Specifically, since each target verification image is marked with a unique identifier, the target verification image corresponding to the unique identifier of the verified document to be verified can be found from the target verification images corresponding to one or more to-be-verified images, and Use this target verification image as the target archive image. Then, based on the target archive image, archive information corresponding to this business transaction can be generated. When handling business, it is necessary to archive the images corresponding to the documents to be verified. For example, in the scenario of invoice financing, archiving the images corresponding to the invoices can make the business processing information more complete and help to deal with the inaccuracies of external financial regulatory agencies. Regular spot checks.

In the embodiment of the present application, optionally, "cropping the image to be verified to obtain a sub-image to be verified corresponding to each document to be verified" in step 204 includes: The positions corresponding to the second text combination and the third text combination are determined sequentially in the verification image, wherein the second text combination is the universal ending text combination of the document to be verified, and the third text combination is the document to be verified. common starting text combinations; based on each The position corresponding to the second word combination, the position corresponding to the third word combination, and the preset segmentation ratio are used to determine the target cropping position; the image to be verified is cropped based on the target cropping position, and the image to be verified is obtained. The sub-image to be verified corresponding to the document to be verified.

In this embodiment, when cropping the image to be verified, the position of the second word combination and the position of the third word combination can be determined from the image to be verified in the order of the second word combination - the third word combination. Location. Here, the second text combination can be a common ending text combination of the document to be verified. For example, when the document to be verified is an invoice, the common ending text combination of each invoice can be "Payee", "Review", "Invoicer" ", "Seller", etc., these text combinations are usually in the last line of the invoice; the third text combination can be the common starting text combination of the document to be verified, for example, when the document to be verified is an invoice, the common starting text combination of each invoice The text combination can be "invoice code", "VAT special invoice", "VAT general invoice", etc. These text combinations are usually on the first line of the invoice. After determining the position corresponding to the second text combination and the position corresponding to the third text combination, you can use the position corresponding to the second text combination, the position corresponding to the third text combination, and the preset division ratio to determine the two to-be-used characters. Verify target crop position between documents. Here, the preset split ratio can be calculated in advance. For example, the second text combination is located on the last line of the document to be verified, and the third text combination is located on the first line of the document to be verified, then it can be divided between the two documents to be verified. When the distances between the documents are different, the optimal split ratio is calculated respectively, and the preset split ratio is determined based on multiple optimal split ratios to ensure that when two consecutive documents to be verified are split according to the preset split ratio, the two consecutive documents to be verified can be split. The documents to be verified are complete. After the target cropping position is determined, the image to be verified can be cropped according to the target cropping position, and then the sub-image to be verified corresponding to each document to be verified can be obtained. This application determines the position of the second text combination in the image to be verified, the position of the third text combination in the image to be verified, and the preset division ratio, and then analyzes the image to be verified that contains multiple documents to be verified. By cropping, you can simply and conveniently determine the sub-image to be verified corresponding to each document to be verified, making it easier to accurately find the image for archiving later.

If the image to be verified contains two documents to be verified, then the segmentation can be performed by determining the positions of a second text combination and a third text combination in the image to be verified. If the image to be verified contains three documents to be verified, To verify the document, you only need to determine the positions of the two sets of second text combinations and third text combinations in the image to be verified to perform segmentation, and so on. Therefore, after determining the number of documents to be verified corresponding to each image to be verified, the positions of the second text combination and the third text combination can be determined from the text information of the image to be verified based on this number. When the above-mentioned After the positions of the second word combination and the third word combination are located, the position of the second word combination and the third word combination can no longer be determined. The embodiment of the present application uses the number of documents to be verified. When the number of determined positions of the second character combination and the third character combination is equal to the number of documents to be verified, it can be stopped, which can effectively reduce the time for position determination. , improve the efficiency of position determination.

In the embodiment of the present application, optionally, before "obtaining at least one image to be verified" in step 201, the method further includes: obtaining a list of document numbers, wherein the list of document numbers includes each of the The document number corresponding to the document to be verified; accordingly, the target query information includes the target document number; after step 205, the method further includes: identifying the target document number in the target query information, and The target document number is eliminated from the document number list to obtain a missing document list.

In this embodiment, before acquiring one or more images to be verified, a list of document numbers can first be acquired. Here, the document number list may include the document number corresponding to each document to be verified, where the document number and the document to be verified are also in one-to-one correspondence. When the document to be verified is an invoice, the document number can be the invoice number. When the target query information includes the target document number, if the target query information corresponding to each document to be verified has been determined from the text information, the target document number can be determined from the target query information, and the target document number can be changed from the document number Eliminate it from the list so that the target document number is no longer included in the document number list, so that when the target query information corresponding to each document to be verified contains After all the target document numbers are eliminated from the document number list, the missing document list can be obtained. The document numbers included in the missing document list can be the document numbers corresponding to the documents to be verified that failed to recognize the text information. Subsequently, you can directly determine which documents to be verified have not been identified based on the list of missing documents, and then you can find these documents to be verified and re-identify them. Compared with manually finding the missing documents to be verified from many documents to be verified, and then re-identifying them. By re-identifying these missed documents to be verified, this application can greatly reduce the manual workload. In the embodiment of this application, before performing image text recognition to be verified, a document number list containing the document numbers corresponding to all documents to be verified is first obtained, and then the document numbers corresponding to the identified documents to be verified are removed from the document number list, and finally the document numbers are The remaining document numbers in the number list are the document numbers corresponding to the documents to be verified that cannot be identified or cannot be identified accurately during the OCR process. Especially when the data volume of the documents to be verified is particularly large, the documents to be verified can be quickly and accurately located. Verification documents are missed, improving the efficiency of determining missing documents to be verified.

Further, as a specific implementation of the method in Figure 1, an embodiment of the present application provides a business processing device based on image recognition, as shown in Figure 3. The device includes:

Optionally, the device also includes:

A quantity determination module, configured to determine the number of documents to be verified contained in the image to be verified based on the text information after obtaining the text information corresponding to each of the images to be verified;

A marking module, configured to determine a unique identifier from the text information when it contains one of the documents to be verified, mark the image to be verified through the unique identifier, and use the marked image to be verified as the target verification image; When multiple documents to be verified are included, the image to be verified is cropped to obtain a sub-image to be verified corresponding to each document to be verified, and the content of each document to be verified is determined from the text information. A unique identifier corresponding to the sub-image to be verified, marking the sub-image to be verified by the unique identifier, and using each marked sub-image to be verified as the target verification image;

Correspondingly, the archiving module is configured to: determine the target verification image corresponding to any of the documents to be verified based on the unique identification from the target verification image corresponding to the at least one image to be verified, Archive the image as the target.

Optionally, the quantity determination module is configured to: identify the number of occurrences of the first text combination from the text information, and use the number of occurrences as the number of documents to be verified contained in the image to be verified, Wherein, the first text combination is a text combination that exists in each of the documents to be verified and appears only once.

Optionally, the marking module includes:

The first position determination unit is used to sequentially determine the positions corresponding to the second text combination and the third text combination from the image to be verified, wherein the second text combination is a universal ending text combination of the document to be verified, The third text combination is a common starting text combination of the document to be verified;

The second position determination unit is used to determine the target cropping position based on the position corresponding to each group of the second character combination, the position corresponding to the third character combination, and the preset segmentation ratio;

A cropping unit, configured to crop the image to be verified based on the target cropping position to obtain the sub-image to be verified corresponding to each of the documents to be verified.

Optionally, when the image to be verified contains multiple documents to be verified, the verification module is configured to: divide the text information into sub-text information corresponding to each of the target verification images, The target query information corresponding to the document to be verified is extracted from each sub-text information respectively.

Optionally, the verification module includes:

An interface calling unit, configured to call a preset document verification interface, input the target query information into the preset document verification interface, and receive the document status information returned by the preset document verification interface, wherein the document status Information includes registered status and unregistered status;

A judgment unit configured to determine that the document to be verified has not passed verification when the document status information is in a registered state; and to determine that the document to be verified has passed verification when the document status information is in an unregistered state.

Optionally, the device also includes:

A list acquisition module, configured to obtain a list of document numbers before acquiring at least one image to be verified, wherein the list of document numbers includes a document number corresponding to each of the documents to be verified;

Correspondingly, the target query information includes the target document number; the device further includes:

An elimination module, configured to identify the target document number in the target query information after determining the target query information corresponding to each document to be verified from the text information, and add the target document number to the target query information. Eliminate from the list of document numbers to obtain a list of missing documents.

It should be noted that for other corresponding descriptions of each functional unit involved in the image recognition-based business processing device provided by the embodiment of the present application, reference can be made to the corresponding descriptions in the methods of Figures 1 to 2, and will not be described again here.

Based on the above methods shown in Figures 1 to 2, correspondingly, embodiments of the present application also provide a storage medium on which a computer program is stored. When the computer program is executed by a processor, the above described Figures 1 to 2 are implemented. The business processing method based on image recognition is shown.

Based on this understanding, the technical solution of the present application can be embodied in the form of a software product. The software product can be stored in a computer-readable storage medium. The computer-readable storage medium can be a non-volatile storage medium (can be a CD). -ROM, U disk, mobile hard disk, etc.), or it can be a volatile storage medium. The computer-readable storage medium includes a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) Execute the methods described in each implementation scenario of this application.

Based on the above methods shown in Figures 1 to 2 and the virtual device embodiment shown in Figure 3, in order to achieve the above purpose, embodiments of the present application also provide a computer device, which can be a personal computer, a server, a network Equipment, etc., the computer equipment includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to implement the above-mentioned business processing method based on image recognition as shown in Figures 1 to 2.

Optionally, the computer device may also include a user interface, a network interface, a camera, a radio frequency (Radio Frequency, RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), etc. The optional user interface may also include a USB interface, a card reader interface, etc. Optional network interfaces may include standard wired interfaces, wireless interfaces (such as Bluetooth interfaces, WI-FI interfaces), etc.

Those skilled in the art can understand that the structure of a computer device provided in this embodiment does not constitute a limitation on the computer device, and may include more or less components, or combine certain components, or arrange different components.

The storage medium may also include an operating system and a network communication module. An operating system is a program that manages and saves the hardware and software resources of a computer device and supports the operation of information processing programs and other software and/or programs. The network communication module is used to implement communication between components within the storage medium, as well as communication with other hardware and software in the physical device.

Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform, or can also be implemented by hardware. First, one or more images to be verified can be obtained, and each image to be verified can contain one or more documents to be verified. After obtaining one or more images to be verified, text recognition can be performed on each image to be verified, and the text information corresponding to each image to be verified can be obtained. After obtaining the text information corresponding to each image to be verified, when the image to be verified contains a document to be verified, the target query information corresponding to the document to be verified can be determined directly from the text information corresponding to the image to be verified; when the document to be verified is to be When the image contains multiple documents to be verified, the target query information corresponding to each document to be verified can be determined from the text information corresponding to the image to be verified. After the target query information is determined, each document to be verified can be verified based on the target query information corresponding to the document to be verified, and it is determined whether the document to be verified can pass the verification. If there is a document to be verified that has passed the verification, the document to be verified that has passed the verification can be used to handle related business in the future. During the business processing process, in order to complete business-related information, documents to be verified can also be archived. Specifically, the target archive image corresponding to the verified document to be verified can be found from the obtained image to be verified, and then the archive information of the business can be generated based on the target archive image. The embodiments of this application can simplify the verification process of documents to be verified, greatly improve the efficiency of verification and archiving, and help improve the user experience.

Those skilled in the art can understand that the accompanying drawing is only a schematic diagram of a preferred implementation scenario, and the modules or processes in the accompanying drawing are not necessarily necessary for implementing the present application. Those skilled in the art can understand that the modules in the devices in the implementation scenario can be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more devices different from the implementation scenario. The modules of the above implementation scenarios can be combined into one module or further split into multiple sub-modules.

The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. What is disclosed above are only a few specific implementation scenarios of the present application. However, the present application is not limited thereto. Any changes that can be thought of by those skilled in the art should fall within the protection scope of the present application.

Claims

A business processing method based on image recognition, wherein the method includes:

Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each of the images to be verified, wherein each of the images to be verified includes at least one document to be verified;

Determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

After any of the documents to be verified passes the verification, a target archive image corresponding to any of the documents to be verified is determined based on the at least one image to be verified, and the business is generated based on the target archive image archived information.
The method according to claim 1, wherein after obtaining the text information corresponding to each of the images to be verified, the method further includes:

Based on the text information, determine the number of documents to be verified contained in the image to be verified;

When a document to be verified is included, a unique identifier is determined from the text information, the image to be verified is marked with the unique identifier, and the marked image to be verified is used as the target verification image;

When multiple documents to be verified are included, the image to be verified is cropped to obtain a sub-image to be verified corresponding to each document to be verified, and the content of each document to be verified is determined from the text information. A unique identifier corresponding to the sub-image to be verified, marking the sub-image to be verified by the unique identifier, and using each marked sub-image to be verified as the target verification image;

Correspondingly, determining the target archive image corresponding to any of the documents to be verified based on the at least one image to be verified includes:

From the target verification image corresponding to the at least one image to be verified, the target verification image corresponding to any of the documents to be verified is determined based on the unique identifier as the target archive image.
The method according to claim 2, wherein determining the number of documents to be verified contained in the image to be verified based on the text information includes:

The number of occurrences of the first character combination is identified from the text information, and the number of occurrences is used as the number of documents to be verified contained in the image to be verified, wherein the first character combination is for each of the documents to be verified. Verify the text combination that exists in the document and appears only once.
The method according to claim 2, wherein the cropping of the image to be verified to obtain the sub-image to be verified corresponding to each of the documents to be verified includes:

The positions corresponding to the second text combination and the third text combination are determined sequentially from the image to be verified, wherein the second text combination is the universal ending text combination of the document to be verified, and the third text combination is the Describe the common starting text combination of the document to be verified;

Determine the target cropping position based on the position corresponding to the second text combination of each group, the position corresponding to the third text combination, and the preset segmentation ratio;

The image to be verified is cropped based on the target cropping position to obtain a sub-image to be verified corresponding to each document to be verified.
The method according to claim 2, wherein when the image to be verified contains multiple documents to be verified, the target query information corresponding to each document to be verified is determined from the text information. ,include:

The text information is divided into sub-text information corresponding to each target verification image, and target query information corresponding to the document to be verified is extracted from each sub-text information.
The method according to claim 1, wherein determining whether the document to be verified passes verification based on the target query information includes:

Call the preset document verification interface, input the target query information into the preset document verification interface, and receive the document status information returned by the preset document verification interface, where the document status information includes registered status and Unregistered status;

When the document status information is in the registered state, it is determined that the document to be verified has not passed verification;

When the document status information is in the unregistered state, it is determined that the document to be verified passes the verification.
The method according to claim 1, wherein before obtaining at least one image to be verified, the method further includes:

Obtain a list of document numbers, wherein the list of document numbers includes the document number corresponding to each of the documents to be verified;

Correspondingly, the target query information includes a target document number; after determining the target query information corresponding to each of the documents to be verified from the text information, the method further includes:

Identify the target document number in the target query information, and remove the target document number from the document number list to obtain a missing document list.
A business processing device based on image recognition, which includes:

A text recognition module, configured to obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each image to be verified, wherein each image to be verified includes at least one Documents to be verified;

A verification module, configured to determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

An archiving module, configured to determine a target archive image corresponding to any of the documents to be verified based on the at least one image to be verified after the document to be verified passes the verification, and based on the target archive image , generate archive information of the business.
A storage medium on which a computer program is stored, wherein the following steps are implemented when the computer program is executed by a processor:

Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each of the images to be verified, wherein each of the images to be verified includes at least one document to be verified;

Determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

After any of the documents to be verified passes the verification, a target archive image corresponding to any of the documents to be verified is determined based on the at least one image to be verified, and the business is generated based on the target archive image archived information.
The storage medium according to claim 9, wherein, after performing the step of obtaining the text information corresponding to each of the images to be verified, it is further used to perform:

Based on the text information, determine the number of documents to be verified contained in the image to be verified;

When a document to be verified is included, a unique identifier is determined from the text information, the image to be verified is marked with the unique identifier, and the marked image to be verified is used as the target verification image;

When multiple documents to be verified are included, the image to be verified is cropped to obtain a sub-image to be verified corresponding to each document to be verified, and the content of each document to be verified is determined from the text information. A unique identifier corresponding to the sub-image to be verified, marking the sub-image to be verified by the unique identifier, and using each marked sub-image to be verified as the target verification image;

Correspondingly, when performing the step of determining the target archive image corresponding to any of the documents to be verified based on the at least one image to be verified, specifically execute:

From the target verification image corresponding to the at least one image to be verified, the target verification image corresponding to any of the documents to be verified is determined based on the unique identifier as the target archive image.
The storage medium according to claim 10, wherein when determining the number of documents to be verified contained in the image to be verified based on the text information, the specific execution is:

The number of occurrences of the first character combination is identified from the text information, and the number of occurrences is used as the number of documents to be verified contained in the image to be verified, wherein the first character combination is for each of the documents to be verified. Verify the text combination that exists in the document and appears only once.
The storage medium according to claim 10, wherein when performing the cropping process on the image to be verified to obtain the sub-image to be verified corresponding to each of the documents to be verified, the specific execution is:

The positions corresponding to the second text combination and the third text combination are determined sequentially from the image to be verified, wherein the second text combination is the universal ending text combination of the document to be verified, and the third text combination is the Describe the common starting text combination of the document to be verified;

Determine the target cropping position based on the position corresponding to the second text combination of each group, the position corresponding to the third text combination, and the preset segmentation ratio;

The image to be verified is cropped based on the target cropping position to obtain a sub-image to be verified corresponding to each document to be verified.
The storage medium according to claim 10, wherein when the image to be verified contains a plurality of the documents to be verified, determining the target corresponding to each of the documents to be verified from the text information is performed. When querying information, specifically execute:

The text information is divided into sub-text information corresponding to each target verification image, and target query information corresponding to the document to be verified is extracted from each sub-text information.
The storage medium according to claim 9, wherein when determining whether the document to be verified passes verification based on the target query information, the specific execution is:

Call the preset document verification interface, input the target query information into the preset document verification interface, and receive the document status information returned by the preset document verification interface, where the document status information includes registered status and Unregistered status;

When the document status information is in the registered state, it is determined that the document to be verified has not passed verification;

When the document status information is in the unregistered state, it is determined that the document to be verified passes the verification.
A computer device includes a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the following steps when executing the computer program:

Obtain at least one image to be verified, perform text recognition on the at least one image to be verified, and obtain text information corresponding to each of the images to be verified, wherein each of the images to be verified includes at least one document to be verified;

Determine the target query information corresponding to each of the documents to be verified from the text information, and determine whether the document to be verified passes the verification based on the target query information;

After any of the documents to be verified passes the verification, a target archive image corresponding to any of the documents to be verified is determined based on the at least one image to be verified, and the business is generated based on the target archive image archived information.
The computer device according to claim 15, wherein, after performing the step of obtaining the text information corresponding to each of the images to be verified, it is further configured to perform:

Based on the text information, determine the number of documents to be verified contained in the image to be verified;

When a document to be verified is included, a unique identifier is determined from the text information, the image to be verified is marked with the unique identifier, and the marked image to be verified is used as the target verification image;

When multiple documents to be verified are included, the image to be verified is cropped to obtain a sub-image to be verified corresponding to each document to be verified, and the content of each document to be verified is determined from the text information. A unique identifier corresponding to the sub-image to be verified, marking the sub-image to be verified by the unique identifier, and using each marked sub-image to be verified as the target verification image;

Correspondingly, when performing the step of determining the target archive image corresponding to any of the documents to be verified based on the at least one image to be verified, specifically execute:

From the target verification image corresponding to the at least one image to be verified, the target verification image corresponding to any of the documents to be verified is determined based on the unique identifier as the target archive image.
The computer device according to claim 16, wherein when determining the number of documents to be verified contained in the image to be verified based on the text information, the specific execution is:

The number of occurrences of the first character combination is identified from the text information, and the number of occurrences is used as the number of documents to be verified contained in the image to be verified, wherein the first character combination is for each of the documents to be verified. Verify the text combination that exists in the document and appears only once.
The computer device according to claim 16, wherein when performing the cropping process of the image to be verified to obtain the sub-image to be verified corresponding to each of the documents to be verified, the specific execution is:

The positions corresponding to the second text combination and the third text combination are determined sequentially from the image to be verified, wherein the second text combination is the universal ending text combination of the document to be verified, and the third text combination is the Describe the common starting text combination of the document to be verified;

Determine the target cropping position based on the position corresponding to the second text combination of each group, the position corresponding to the third text combination, and the preset segmentation ratio;

The image to be verified is cropped based on the target cropping position to obtain a sub-image to be verified corresponding to each document to be verified.
The computer device according to claim 16, wherein when the image to be verified contains a plurality of the documents to be verified, performing the step of determining the target corresponding to each of the documents to be verified from the text information. When querying information, specifically execute:

The text information is divided into sub-text information corresponding to each target verification image, and target query information corresponding to the document to be verified is extracted from each sub-text information.
The computer device according to claim 15, wherein when determining whether the document to be verified passes verification based on the target query information, the specific execution is:

Call the preset document verification interface, input the target query information into the preset document verification interface, and receive the document status information returned by the preset document verification interface, where the document status information includes registered status and Unregistered status;

When the document status information is in the registered state, it is determined that the document to be verified has not passed verification;

When the document status information is in the unregistered state, it is determined that the document to be verified passes the verification.