Invoice online verification method, device and medium

Abstract

The application discloses an invoice online verification method, device and medium, the method comprises the following steps: obtaining an invoice verification request; identifying the invoice through an invoice identification neural network model to obtain identified invoice information; determining invoice authenticity key information to be verified in the invoice information; calling a tax bureau inspection interface and matching the invoice authenticity key information in the database of the tax bureau; if the matching is successful, it is determined that the invoice authenticity key information verification is passed; determining the file type of the invoice; the file type includes a PDF invoice or an OFD invoice; determining the seal verification model of the invoice according to the file type; and performing seal verification on the invoice according to the seal verification model to determine whether the invoice is tampered with. The authenticity of the invoice can be verified, and it is determined whether the invoice is tampered with, the invoice is double-verified, the false invoice or the real invoice but the tampered invoice is accurately found, and the invoice verification accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, and in particular to an invoice online verification method, device and medium. BACKGROUND

[0002] At present, in order to adapt to the economic and social development and the needs of tax modernization construction, the electronic value-added tax invoice is gradually implemented. Invoice electronicization is an important measure to improve the convenience of taxpayers, save resource costs, improve enterprise management level and enhance the modern tax governance ability.

[0003] For ERP software, various enterprises begin to use electronic invoices for reimbursement, and need to verify the authenticity and security of electronic invoices. However, the method for verifying the authenticity and security of electronic invoices is relatively single and simple, resulting in low accuracy of invoice verification. SUMMARY

[0004] The embodiments of the present application provide an invoice online verification method, device and medium, which are used to solve the problem of low accuracy of invoice verification.

[0005] The embodiments of the present application adopt the following technical solutions:

[0006] On the one hand, the embodiments of the present application provide an invoice online verification method, which comprises the following steps: obtaining a verification request of an invoice; identifying the invoice by using an invoice identification neural network model to obtain identified invoice information; determining invoice authenticity key information to be verified in the invoice information; calling a tax bureau inspection interface, and matching the invoice authenticity key information in the database of the tax bureau; if the matching is successful, it is determined that the invoice authenticity key information verification is passed; determining the file type of the invoice; the file type includes a PDF invoice or an OFD invoice; determining a signature verification model of the invoice according to the file type; and performing signature verification on the invoice according to the signature verification model to determine whether the invoice is tampered with.

[0007] In one example, the signature verification on the invoice according to the signature verification model specifically comprises: if the file type is a PDF invoice, the invoice information is encrypted according to a pre-set PDF file encryption algorithm to obtain a PDF file encryption result of the invoice; the signature value of the invoice is read; the signature value is compared with the PDF file encryption result; if the signature value is consistent with the PDF file encryption result, it is determined that the invoice is not tampered with; and if the signature value is inconsistent with the PDF file encryption result, it is determined that the invoice is tampered with.

[0008] In one example, the signature verification of the invoice according to the signature verification model specifically includes: if the file type is an OFD invoice, encrypting the invoice information according to a pre-set OFD file encryption algorithm to obtain an OFD file encryption result of the invoice; reading a signature value of the invoice; comparing the signature value with the OFD file encryption result; if the signature value is consistent with the OFD file encryption result, determining that the invoice is not tampered with; and if the signature value is inconsistent with the OFD file encryption result, determining that the invoice is tampered with.

[0009] In one example, a sample invoice is obtained; the sample invoice is taken as input, and invoice information of the sample invoice is taken as a label to perform supervised training on an initial neural network model; and the trained initial neural network model is verified, and after verification, the trained initial neural network model is determined as the invoice identification neural network model.

[0010] In one example, if the invoice authenticity key information matching fails, it is determined whether the invoice information includes invoice standard information; the invoice standard information is used to represent a standard template element constituting a standard general invoice; if not, the invoice is determined as a fake invoice; and if yes, fake invoice verification notification information of the invoice is generated according to the unmatched invoice authenticity key information; and the fake invoice verification notification information is sent to a bound fake invoice verification terminal device.

[0011] In one example, after the fake invoice verification notification information is sent to the fake invoice verification terminal device, the method further includes: receiving feedback information of the fake invoice verification terminal device; if the feedback information is fake invoice verification failure, performing secondary identification on the invoice through the invoice identification neural network model to obtain secondary identified invoice information; determining invoice authenticity key information to be subjected to secondary invoice authenticity verification in the secondary identified invoice information; calling the tax bureau inspection interface to match the secondary invoice authenticity verification invoice authenticity key information in the database of the tax bureau; if the matching is successful, determining that the secondary invoice authenticity verification invoice authenticity key information passes the verification; if the matching fails, obtaining verification logs of the invoice and sending the verification logs to an administrator terminal device; and the administrator terminal device is different from the fake invoice verification terminal device.

[0012] In one example, after the verification logs are sent to the administrator terminal device, the method further includes: obtaining log feedback information of the administrator terminal device; determining whether the invoice is a fake invoice according to the log feedback information; if not, updating the invoice identification neural network model based on the invoice information; and if yes, determining that the fake invoice verification passes.

[0013] In one example, the method further comprises: after the pseudo-invoice verification passes, generating pseudo-invoice notification information of the invoice according to the invoice authenticity key information; and sending the pseudo-invoice notification information to the invoice verification request terminal device.

[0014] In another aspect, the embodiments of the present application provide an invoice online verification device, comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: acquire a verification request of an invoice; identify the invoice by an invoice identification neural network model to obtain identified invoice information; determine invoice authenticity key information to be verified for invoice authenticity in the invoice information; call a tax bureau inspection interface to match the invoice authenticity key information in a database of the tax bureau; if the matching is successful, determine that the invoice authenticity key information passes the verification; determine a file type of the invoice; the file type includes a PDF invoice or an OFD invoice; determine a seal verification model of the invoice according to the file type; and perform seal verification on the invoice according to the seal verification model to determine whether the invoice is tampered with.

[0015] In another aspect, the embodiments of the present application provide an invoice online verification nonvolatile computer storage medium, which stores computer executable instructions, and the computer executable instructions are configured to: acquire a verification request of an invoice; identify the invoice by an invoice identification neural network model to obtain identified invoice information; determine invoice authenticity key information to be verified for invoice authenticity in the invoice information; call a tax bureau inspection interface to match the invoice authenticity key information in a database of the tax bureau; if the matching is successful, determine that the invoice authenticity key information passes the verification; determine a file type of the invoice; the file type includes a PDF invoice or an OFD invoice; determine a seal verification model of the invoice according to the file type; and perform seal verification on the invoice according to the seal verification model to determine whether the invoice is tampered with.

[0016] The above at least one technical scheme adopted by the embodiments of the present application can achieve the following beneficial effects:

[0017] The invoice authenticity can be verified by calling the inspection interface of the tax bureau, and after the authenticity verification passes, the invoice is further verified by the seal verification model to determine whether the invoice is tampered with, so as to realize double verification of the invoice, accurately find fake invoices or real invoices that are tampered with, and improve the invoice verification accuracy on the premise of ensuring the invoice verification efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the present application, some embodiments of the present application will be described in detail below with reference to the accompanying drawings, in which:

[0019] Figure 1 A flowchart of an invoice online verification method provided by an embodiment of the present application;

[0020] Figure 2 An example diagram of an invoice online verification provided by an embodiment of the present application;

[0021] Figure 3 A structural diagram of an invoice online verification device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be described clearly and completely below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0023] Some embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0024] Figure 1 A flowchart of an invoice online verification method provided by an embodiment of the present application. The method can be applied to different business fields, such as the Internet financial business field, the e-commerce business field, the instant messaging business field, the game business field, the public service business field, etc. Some input parameters or intermediate results in the flowchart allow manual intervention to adjust to help improve accuracy.

[0025] The implementation of the analysis method related by the embodiments of the present application can be a terminal device or a server, and the present application does not make special limitations thereon. For the convenience of understanding and description, the following embodiments are described in detail with the server as an example.

[0026] It should be noted that the server can be a single device, or a system composed of multiple devices, i.e., a distributed server, and the present application does not make specific limitations thereon.

[0027] It should be noted that the invoice verification is directly integrated into the business system, without the need to install additional verification functions, and through the true and false checking operation and the signature verification operation, the multi-way online verification of the invoice is completed.

[0028] In the image viewing interface of the business system, the invoice file to be verified is selected, and the invoice authenticity verification and seal verification call the verification service on the application server. Users do not need to install and understand the complex procedures of the client desktop software, and can directly enter the business system to select the corresponding invoice and complete the multi-channel verification of the invoice through simple operation.

[0029] In addition, based on the B / S (browser / server) architecture, the B end refers to a browser capable of parsing and executing JavaScript, and the S end refers to a verification service on an application server. The browser on the PC end directly communicates with the service on the application server. The service includes two aspects: communication with the tax bureau service and local analysis and verification of the seal.

[0030] Figure 1 The process in the method can include the following steps:

[0031] S101: Obtain a verification request for an invoice.

[0032] That is, an invoice image uploaded by a user is obtained.

[0033] S102: Identify the invoice by using an invoice identification neural network model to obtain identified invoice information.

[0034] In some embodiments of the present application, first, a sample invoice needs to be obtained. Then, the sample invoice is taken as input, and the invoice information of the sample invoice is taken as a label to perform supervised training on an initial neural network model. Finally, the trained initial neural network model is verified, and after verification, the trained initial neural network model is determined as the invoice identification neural network model.

[0035] S103: In the invoice information, determine invoice authenticity key information to be verified for invoice authenticity.

[0036] S104: Call a tax bureau verification interface to match the invoice authenticity key information in the database of the tax bureau.

[0037] The invoice authenticity key information is used to represent the authenticity of the invoice, for example, the invoice authenticity key information includes the number, code, and date of the invoice.

[0038] It should be noted that when the invoice is issued, there will be a corresponding invoice record in the database of the tax bureau.

[0039] S105: If the matching is successful, it is determined that the invoice authenticity key information verification is passed.

[0040] In some embodiments of the present application, when the invoice authenticity key information matching fails, it may be that the invoice is a fake invoice, or it may be that the invoice is correct but the identification is wrong due to special reasons, such as the invoice picture is not clear, the neural network model identification is wrong. Therefore, the reason for the failure of the automatic analysis of the invoice authenticity key information is continued.

[0041] Specifically, if the invoice authenticity key information matching fails, it is determined whether the invoice information includes invoice standard information. Among them, the invoice standard information is used to represent the standard template elements constituting the standard general invoice. That is, the standard template elements of the general invoice standard format will be set by the tax bureau, so that the tax bureau can issue standard general invoices based on the general invoice standard format.

[0042] If not, the invoice is determined to be a fake invoice. That is, at this time, the invoice does not meet the standard of the tax bureau to issue an invoice, so it can be determined that it is not issued by the tax bureau, and there is no possibility of correct invoice identification error.

[0043] If yes, the fake invoice verification notification information of the invoice is generated according to the unmatched invoice authenticity key information. That is, when the invoice information includes the invoice standard information, it can be determined that the current invoice meets the format of the tax bureau to issue an invoice, so at this time the invoice may be a fake invoice, or the invoice may be correct but the identification is wrong.

[0044] Therefore, the fake invoice verification notification information is sent to the bound fake invoice verification terminal device. Therefore, the reason is analyzed through the feedback information of the fake invoice verification terminal device.

[0045] Specifically, the feedback information of the fake invoice verification terminal device is received. If the feedback information is fake invoice verification failure.

[0046] That is, when the feedback information is fake invoice verification failure, it means that the fake invoice verification terminal device feedbacks that the invoice is not a fake invoice, so at this time it may be that the invoice is correct but the identification is wrong, so the invoice is identified again.

[0047] Based on this, the invoice is identified again by the invoice identification neural network model to obtain the second identification invoice information. Then, in the second identification invoice information, the invoice authenticity key information to be subjected to the second invoice authenticity verification is determined. Then, the tax bureau inspection interface is called, and the second invoice authenticity key information subjected to the second invoice authenticity verification is matched in the database of the tax bureau.

[0048] If the matching is successful, it is determined that the second invoice authenticity key information subjected to the second invoice authenticity verification is verified.

[0049] That is, when the matching succeeds, it means that the invoice is correct, but the first identification result is wrong. However, when the matching fails, it is considered that the first identification is not wrong, but at the same time, the pseudo-invoice verification terminal device feedbacks that the invoice is not a pseudo-invoice. Since the pseudo-invoice verification terminal device may exist cheating, that is, the invoice is actually a pseudo-invoice, but it feedbacks that the invoice is not a pseudo-invoice.

[0050] Based on this, if the matching fails, the verification log of the invoice is obtained, and the verification log is sent to the administrator terminal device. The administrator terminal device is different from the pseudo-invoice verification terminal device.

[0051] Then, the log feedback information of the administrator terminal device is obtained. Then, according to the log feedback information, it is determined whether the invoice is a pseudo-invoice.

[0052] If not, the invoice identification neural network model is updated based on the invoice information. At this time, it means that the invoice identification neural network model has an error, and the invoice identification neural network model needs to be updated.

[0053] If yes, it is determined that the pseudo-invoice verification is passed. At this time, it means that the pseudo-invoice verification terminal device exists cheating.

[0054] After the pseudo-invoice verification is passed, the pseudo-invoice notification information of the invoice is generated according to the invoice authenticity key information. Therefore, the pseudo-invoice notification information is sent to the invoice verification request terminal device.

[0055] It should be noted that when the invoice authenticity key information verification is passed, S106-S108 is continued to be executed, otherwise S106-S108 is not executed any more.

[0056] S106: Determine the file type of the invoice; the file type includes a PDF invoice or an OFD invoice.

[0057] S107: According to the file type, determine the seal verification model of the invoice.

[0058] The PDF invoice has a corresponding seal verification model, and the OFD invoice has a specific corresponding seal verification model.

[0059] S108: According to the seal verification model, the seal verification of the invoice is performed to determine whether the invoice is tampered.

[0060] In some embodiments of the present application, if the file type is a PDF invoice, the invoice information is encrypted according to a pre-set PDF file encryption algorithm to obtain a PDF file encryption result of the invoice. Then, the seal value of the invoice is read. Then, the seal value is compared with the PDF file encryption result.

[0061] If the signature value is consistent with the PDF file encryption result, it is determined that the invoice has not been tampered with. If the signature value is inconsistent with the PDF file encryption result, it is determined that the invoice has been tampered with.

[0062] It should be noted that the pre-set PDF file encryption algorithm can be set according to actual needs, and the signature value of the invoice is also obtained by encrypting the original invoice information through the PDF file encryption algorithm.

[0063] In addition, if the file type is an OFD invoice, the invoice information is encrypted according to a pre-set OFD file encryption algorithm to obtain an OFD file encryption result of the invoice. Then, the signature value of the invoice is read. The signature value is compared with the OFD file encryption result.

[0064] If the signature value is consistent with the OFD file encryption result, it is determined that the invoice has not been tampered with. If the signature value is inconsistent with the OFD file encryption result, it is determined that the invoice has been tampered with.

[0065] It should be noted that the pre-set OFD file encryption algorithm can be set according to actual needs. And the signature value of the invoice is also obtained by encrypting the original invoice information through the OFD file encryption algorithm.

[0066] It should be noted that when the invoice has not been tampered with, a verification success notification is sent to the verification request terminal, and when the invoice has been tampered with, a signature verification failure notification is sent to the verification request terminal.

[0067] It should be noted that although the embodiments of the present application are introduced and described in sequence with reference to Figure 1 steps S101 to S108, this does not mean that steps S101 to S108 must be executed in strict sequence. The reason why the embodiments of the present application introduce and describe steps S101 to S108 in sequence as shown in Figure 1 is to facilitate the understanding of the technical solutions of the embodiments of the present application by those skilled in the art. In other words, in the embodiments of the present application, the sequence of steps S101 to S108 can be adjusted appropriately according to actual needs.

[0068] By Figure 1 the method, the authenticity of the invoice can be verified by calling the inspection interface of the tax bureau, and after the authenticity verification is passed, the signature verification model is used to verify the signature of the invoice to determine whether the invoice has been tampered with, thereby realizing double verification of the invoice, and under the premise of ensuring the efficiency of the invoice verification, accurately finding fake invoices or real invoices that have been tampered with, and improving the accuracy of the invoice verification.

[0069] More intuitively, Figure 2This is an example diagram of online invoice verification provided in an embodiment of this application.

[0070] The verification server in this application embodiment may include verification service, tax bureau service, and signature verification service.

[0071] exist Figure 2 In this process, a web application initiates a verification request to the verification service. The verification service identifies the invoice, obtains the invoice information, and then sends the invoice information to the State Taxation Administration service. The State Taxation Administration service verifies the authenticity of the invoice based on key authenticity information and sends the verification result back to the verification service. The verification service then sends the original invoice document to the signature verification service, which provides feedback on whether the invoice has been tampered with. Finally, the verification service obtains the verification result of the invoice and sends it back to the web application.

[0072] Based on the same idea, some embodiments of this application also provide devices and non-volatile computer storage media corresponding to the above methods.

[0073] Figure 3 A schematic diagram of an online invoice verification device provided in this application embodiment includes:

[0074] At least one processor; and,

[0075] A memory communicatively connected to the at least one processor; wherein,

[0076] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0077] Verification request to obtain an invoice;

[0078] The invoice is identified using an invoice recognition neural network model to obtain the identified invoice information;

[0079] The invoice information contains key information to determine the authenticity of the invoice that will be verified.

[0080] The system calls the State Taxation Administration's verification interface to match key information regarding the authenticity of the invoices in the State Taxation Administration's database.

[0081] If the match is successful, the key information regarding the authenticity of the invoice has been verified.

[0082] Determine the document type of the invoice; the document type includes PDF invoices or OFD invoices;

[0083] Based on the document type, determine the signature verification model for the invoice;

[0084] According to the signature verification model, the invoice is subjected to signature verification to determine whether the invoice is tampered.

[0085] Some embodiments of the present application provide an invoice online verification nonvolatile computer storage medium, which stores computer executable instructions, the computer executable instructions are configured to:

[0086] Obtain a verification request of an invoice;

[0087] Identify the invoice by an invoice identification neural network model to obtain identified invoice information;

[0088] In the invoice information, determine invoice authenticity key information to be subjected to invoice authenticity verification;

[0089] Call a tax bureau inspection interface, and match the invoice authenticity key information in a database of the tax bureau;

[0090] If the matching is successful, it is determined that the invoice authenticity key information passes the verification;

[0091] Determine the file type of the invoice; the file type includes a PDF invoice or an OFD invoice;

[0092] According to the file type, determine a signature verification model of the invoice;

[0093] According to the signature verification model, the invoice is subjected to signature verification to determine whether the invoice is tampered.

[0094] Each of the embodiments of the present application is described in a progressive manner, and the same or similar parts of each of the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments. In particular, the device and medium embodiments are described simply because they are basically similar to the method embodiments, and the relevant parts can be referred to the part of the method embodiment.

[0095] The device and medium provided by the embodiments of the present application are one-to-one corresponding to the method, and therefore, the device and medium also have the similar beneficial technical effects as the method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device and medium will not be described here.

[0096] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk memory, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0097] The present application is described in reference to the flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in one or more of the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in one or more of the flowchart illustrations and / or block diagrams.

[0098] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in one or more of the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in one or more of the flowchart illustrations and / or block diagrams.

[0099] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in one or more of the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in one or more of the flowchart illustrations and / or block diagrams.

[0100] In a typical configuration, a computing device includes one or more processors (CPUs), input / output (I / O) interfaces, network interfaces, and memory.

[0101] The memory can include non-persistent memory, random access memory (RAM), and / or non-volatile memory such as read only memory (ROM) or flash memory, etc. in a computer readable medium. The memory is an example of computer readable media.

[0102] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0103] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0104] The above only describes the embodiments of the present application and is not intended to limit the present application. Those skilled in the art can make various modifications and changes to the present application. Any modification, equivalent replacement, improvement, etc. made within the technical principles of the present application shall fall within the protection scope of the present application.

Claims

1. An online invoice verification method, characterized by, The method comprises: acquiring a verification request of an invoice; identifying the invoice by an invoice identification neural network model to obtain identified invoice information; determining invoice authenticity key information to be verified in the invoice information; calling a tax bureau inspection interface to match the invoice authenticity key information in a database of the tax bureau; if the matching is successful, determining that the invoice authenticity key information is verified; determining a file type of the invoice; the file type comprises a PDF invoice or an OFD invoice; determining a seal verification model of the invoice according to the file type; performing seal verification on the invoice according to the seal verification model to determine whether the invoice is tampered with; the seal verification on the invoice according to the seal verification model specifically comprises: if the file type is a PDF invoice, encrypting the invoice information according to a pre-set PDF file encryption algorithm to obtain a PDF file encryption result of the invoice; reading a seal value of the invoice; comparing the seal value with the PDF file encryption result; if the seal value is consistent with the PDF file encryption result, determining that the invoice is not tampered with; if the seal value is inconsistent with the PDF file encryption result, determining that the invoice is tampered with; the seal verification on the invoice according to the seal verification model specifically comprises: if the file type is an OFD invoice, encrypting the invoice information according to a pre-set OFD file encryption algorithm to obtain an OFD file encryption result of the invoice; reading a seal value of the invoice; comparing the seal value with the OFD file encryption result; if the seal value is consistent with the OFD file encryption result, determining that the invoice is not tampered with; if the seal value is inconsistent with the OFD file encryption result, determining that the invoice is tampered with; the method further comprises: if the invoice authenticity key information fails to match, determining whether the invoice information comprises invoice standard information; the invoice standard information is used to indicate a standard template element constituting a standard general invoice; if not, determining the invoice as a fake invoice; if yes, generating fake invoice verification notification information of the invoice according to the unmatched invoice authenticity key information; sending the fake invoice verification notification information to a bound fake invoice verification terminal device; after the fake invoice verification notification information is sent to the fake invoice verification terminal device, the method further comprises: receiving feedback information of the fake invoice verification terminal device; if the feedback information is fake invoice verification failure, performing secondary identification on the invoice by the invoice identification neural network model to obtain secondary identified invoice information; determining invoice authenticity key information to be verified in the secondary identified invoice information; calling the tax bureau inspection interface to match the secondary invoice authenticity key information in the database of the tax bureau; if the matching is successful, determining that the secondary invoice authenticity key information is verified; If the matching fails, a verification log of the invoice is obtained, and the verification log is sent to an administrator terminal device; the administrator terminal device is different from the pseudo-invoice verification terminal device; After the verification log is sent to the administrator terminal device, the method further includes: Obtaining log feedback information of the administrator terminal device; According to the log feedback information, it is determined whether the invoice is a pseudo-invoice; If not, the invoice recognition neural network model is updated based on the invoice information; If yes, it is determined that the pseudo-invoice verification is passed; The method further includes: After the pseudo-invoice verification is passed, pseudo-invoice notification information of the invoice is generated according to the invoice authenticity key information; The pseudo-invoice notification information is sent to an invoice verification request terminal device.

2. The method of claim 1, wherein, Before the invoice is recognized by the invoice recognition neural network model to obtain recognized invoice information, the method further includes: Obtaining a sample invoice; The sample invoice is taken as input, and invoice information of the sample invoice is taken as a label to perform supervised training on an initial neural network model; The trained initial neural network model is verified, and after the verification is passed, the trained initial neural network model is determined as the invoice recognition neural network model.

3. An invoice online verification apparatus, characterized by, Comprise: At least one processor; And The memory is in communication connection with the at least one processor; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the invoice online verification method in any one of claims 1-2.

4. An invoice online verification non-transitory computer storage medium storing computer executable instructions, wherein the method comprises: The computer executable instructions are configured to perform the invoice online verification method in any one of claims 1-2.

Images