A method and system for identifying invoice cancellation and cancellation
By entering the invoice information and invoice information in database and fragmenting information, combining optical character recognition technology and information matching module, the existing invoice recognition problem is solved, and the function of quickly identifying wrong invoices and correct invoice information is realized. Improve audit efficiency and simplification of processing processes.
Patent Information
- Application Number
- CN202010841692.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-08-20
- Publication Date
- 2025-02-07
- Estimated Expiration
- 2040-08-20
AI Technical Summary
The existing invoice identification efficiency is low, making it difficult to quickly find the wrong invoice and its corresponding correct invoice information, and it is difficult to distinguish between invalidation and red-blocking.
By entering the invoice information and invoice information into the database, and performing information fragmentation processing, optical character recognition technology is used to scan and identify the invoice, combining the information matching module for efficient matching, automatically identify the validity and month of the invoice, and performing invalidation or red-blocking treatment.
It realizes the rapid identification of wrong invoices and finds the correct invoice information, improves the efficiency of invoice identification, simplifies the process of invalidation and red rushing, and improves the review efficiency.
Smart Images

Figure CN112016422B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of invoice recognition, and in particular relates to a method and system for recognizing the cancellation and cancellation of invoices. Background Art
[0002] Invoice refers to a document issued by the seller to the buyer in economic activities, which includes the name, quality and agreed price of the product or service provided to the buyer. In addition to the advance payment, the invoice must have the elements of payment from the buyer to the seller according to the agreed terms, and must include the date and quantity, which is an important document for accounting and finance. The Chinese accounting system stipulates that the valid invoice for the purchase of products or services is called a tax invoice. The vouchers for government fees and levies are called differently in different periods and for different fee collection items, but they are mostly collectively referred to as administrative fee collection receipts. For internal audits and verifications, each invoice must have a unique serial number to prevent duplicate or skipped invoices.
[0003] In simple terms, an invoice is the original voucher for the costs, expenses or revenues incurred. For a company, an invoice is mainly used as the basis for accounting and as a voucher for payment; for employees, an invoice is mainly used for reimbursement.
[0004] Invoice cancellation means that the operator cancels the invoice that has been issued if the invoice information is incorrectly entered; invoice red indicates that the original invoice is incorrect or needs to be corrected, and a new invoice needs to be issued to adjust the account. The most troublesome part is to find the incorrect invoice and the corresponding correct invoice information. Summary of the invention
[0005] The purpose of the present invention is to overcome the problems and defects of low efficiency in existing invoice recognition, and to provide a method and system for identifying invoice cancellation and redemption, which can quickly find erroneous invoices, find the corresponding correct invoicing information, and intelligently distinguish whether to cancel or redeem an invoice.
[0006] To achieve the above objectives, the technical solution of the present invention is as follows:
[0007] A method for identifying invoice cancellation and redemption, the method comprising the following steps:
[0008] S1: Enter all invoices that need to be identified and establish an invoice information database A. After the invoices are entered, the information is fragmented into N fields, where N is a positive integer greater than or equal to 1;
[0009] S2: Enter the invoicing information corresponding to all invoices that need to be identified and establish an invoicing information database B;
[0010] S3: put any invoice a in the invoice information database A into the invoice information database B for matching;
[0011] S4: Check the matching degree between invoice a and invoice information database B. If there is a piece of invoice information with a matching degree of 100% with invoice a, the invoice content of invoice a is correct, and then invoice a is put into valid invoice subset database C. If there is no piece of invoice information with a matching degree of 100% with invoice a, the invoice content of invoice a is incorrect, and then invoice a is put into invalid invoice subset database D.
[0012] S5: put any invoice b in the invalid invoice subset library D into the invoicing information data B again for matching, and obtain the matching degree. According to the matching degree from (N-1) / N to 1 / N from high to low, check the invoicing information corresponding to each matching degree. The number of invoicing information corresponding to each matching degree is recorded as m, where m is a natural number.
[0013] Select m invoice information with a matching degree of (Nn) / N, and put the m invoices into the invoice information database A for matching. The matching degree is (Nn) / N, and there are q invoice information, q is a positive integer. Among the q invoice information, one is invoice b, and the invoice information is the correct invoice information of invoice b. Then the correct invoice information corresponding to invoice b is recorded in the invoice information subset database E;
[0014] Select m invoice information with a matching degree of (Nn) / N, and put the m invoices into the invoice information database A for matching. If the matching degree is 100%, select invoice information with matching degrees of (Nn-1) / N, (Nn-2) / N to 1 / N in turn and repeat the above steps until the correct invoice information corresponding to invoice b is found and recorded in the invoice information subset database E;
[0015] S6: Identify the invoice month for the invoices in the invalid invoice subset library D. If the year and month on the invoice are consistent with the year and month on the computer system, the invoice is an invoice for the current month. If any of the year and month numbers on the invoice is less than the corresponding year and month number on the computer system, the invoice is not an invoice for this month.
[0016] S7: Find the corresponding invoicing information in the invoicing information subset library E for the invoice of the current month obtained in step S6, and cancel it, and cancel the non-monthly invoice obtained in step S6.
[0017] The invoice described in step S1 is processed into information fragments and divided into four fields, namely, the purchaser information segment, the seller information segment, the tax total information segment, and the goods or taxable services, service name information segment.
[0018] The invoicing information in step S2 includes sales contracts and purchase contracts.
[0019] The valid invoice subset library C and the invalid invoice subset library D in step S4 are subset libraries of the invoice information database A.
[0020] In step S5, n is greater than or equal to 1 and less than or equal to N-1, n is a positive integer, and the selection of the matching degree starts from (N-1) / N.
[0021] In the step S2, the invoicing information entry content corresponds to the information segment of the invoice.
[0022] The identification system includes an invoice scanning and identification module, an invoice information database A module, an invoice information database B module, an information matching module, an invoice valid subset library C module, an invoice invalid subset library D module, an invoice information subset library E module, and an invoice processing module;
[0023] The invoice scanning and recognition module scans and recognizes the invoice through optical character recognition technology, and performs information fragmentation processing on the invoice according to needs;
[0024] The invoice information database A module stores the invoice information obtained from the invoice scanning and recognition module;
[0025] The invoicing information database module B classifies and stores the corresponding invoice information fragments of the sales contract and the purchase contract;
[0026] The information matching module matches the information in the invoice information data module A with the information in the invoicing information database module B;
[0027] The invoice valid subset library C module stores the valid invoice information obtained from the information matching module;
[0028] The invalid invoice subset library D module stores the invalid invoice information obtained from the information matching module;
[0029] The invoicing information subset library E module stores the invoicing information corresponding to the invalid invoices from the information matching module;
[0030] The invoice processing module invalidates or cancels the invoices in the invalid invoice subset library D module, and then reissues a new invoice based on the invoice information in the invoice information subset library E module.
[0031] Compared with the prior art, the present invention has the following beneficial effects:
[0032] The present invention first records the information on the paper invoice digitally, and performs fragmentation processing according to actual needs. The fragmentation processing fragments the complete invoice information. First, it is convenient for storage. The invoice information occupies less memory. The invoice information is effectively summarized, useless information is deleted, and key information is saved. Second, in the matching process, the information errors can be quickly found, which is helpful for the auditor to conduct the audit;
[0033] The invoice information matching method is simple and efficient, which is more than 90% more efficient than the original manual verification. It can check multiple invoices at a time, which is much faster than the original manual verification one by one. At the same time, the matching method is very intuitive, which can clearly tell the auditors where there are errors. At the same time, the verification mode is added to ensure that the invoice information matches the billing information 100%;
[0034] The method also incorporates automatic intelligent review of invoice time, flexibly and effectively distinguishing whether it is an invoice for the current month, and canceling or canceling it according to the month;
[0035] The entire system is also convenient and effective, with a simple input method and high efficiency. The optical character recognition technology has a fast scanning speed and can quickly scan out keywords. The information matching module can quickly match and automatically screen correct or incorrect invoices. The invoice processing module automatically processes incorrect invoices. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 The flowchart of the method for identifying the voided and cancelled invoices of the present invention;
[0037] Figure 2 The figure is a schematic diagram of the structure of the identification system for voiding and canceling invoices in the present invention. DETAILED DESCRIPTION
[0038] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0039] See also Figure 1 and Figure 2 , a method for identifying invoice cancellation and redemption, the identification method comprising the following steps:
[0040] S1: Enter all invoices that need to be identified and establish an invoice information database A. After the invoices are entered, the information is fragmented into N fields, where N is a positive integer greater than or equal to 1;
[0041] S2: Enter the invoicing information corresponding to all invoices that need to be identified and establish an invoicing information database B;
[0042] S3: put any invoice a in the invoice information database A into the invoice information database B for matching;
[0043] S4: Check the matching degree between invoice a and invoice information database B. If there is a piece of invoice information with a matching degree of 100% with invoice a, the invoice content of invoice a is correct, and then invoice a is put into valid invoice subset database C. If there is no piece of invoice information with a matching degree of 100% with invoice a, the invoice content of invoice a is incorrect, and then invoice a is put into invalid invoice subset database D.
[0044] S5: put any invoice b in the invalid invoice subset library D into the invoicing information data B again for matching, and obtain the matching degree. According to the matching degree from (N-1) / N to 1 / N from high to low, check the invoicing information corresponding to each matching degree. The number of invoicing information corresponding to each matching degree is recorded as m, where m is a natural number.
[0045] Select m invoice information with a matching degree of (Nn) / N, and put the m invoices into the invoice information database A for matching. The matching degree is (Nn) / N, and there are q invoice information, q is a positive integer. Among the q invoice information, one is invoice b, and the invoice information is the correct invoice information of invoice b. Then the correct invoice information corresponding to invoice b is recorded in the invoice information subset database E;
[0046] Select m invoice information with a matching degree of (Nn) / N, and put the m invoices into the invoice information database A for matching. If the matching degree is 100%, select invoice information with matching degrees of (Nn-1) / N, (Nn-2) / N to 1 / N in turn and repeat the above steps until the correct invoice information corresponding to invoice b is found and recorded in the invoice information subset database E;
[0047] S6: Identify the invoice month for the invoices in the invalid invoice subset library D. If the year and month on the invoice are consistent with the year and month on the computer system, the invoice is an invoice for the current month. If any of the year and month numbers on the invoice is less than the corresponding year and month number on the computer system, the invoice is not an invoice for this month.
[0048] S7: Find the corresponding invoicing information in the invoicing information subset library E for the invoice of the current month obtained in step S6, and cancel it, and cancel the non-monthly invoice obtained in step S6.
[0049] The invoice described in step S1 is processed into information fragments and divided into four fields, namely, the purchaser information segment, the seller information segment, the tax total information segment, and the goods or taxable services, service name information segment.
[0050] The invoicing information in step S2 includes sales contracts and purchase contracts.
[0051] The valid invoice subset library C and the invalid invoice subset library D in step S4 are subset libraries of the invoice information database A.
[0052] In step S5, n is greater than or equal to 1 and less than or equal to N-1, n is a positive integer, and the selection of the matching degree starts from (N-1) / N.
[0053] In the step S2, the invoicing information entry content corresponds to the information segment of the invoice.
[0054] The identification system includes an invoice scanning and identification module, an invoice information database A module, an invoice information database B module, an information matching module, an invoice valid subset library C module, an invoice invalid subset library D module, an invoice information subset library E module, and an invoice processing module;
[0055] The invoice scanning and recognition module scans and recognizes the invoice through optical character recognition technology, and performs information fragmentation processing on the invoice according to needs;
[0056] The invoice information database A module stores the invoice information obtained from the invoice scanning and recognition module;
[0057] The invoicing information database module B classifies and stores the corresponding invoice information fragments of the sales contract and the purchase contract;
[0058] The information matching module matches the information in the invoice information data module A with the information in the invoicing information database module B;
[0059] The invoice valid subset library C module stores the valid invoice information obtained from the information matching module;
[0060] The invalid invoice subset library D module stores the invalid invoice information obtained from the information matching module;
[0061] The invoicing information subset library E module stores the invoicing information corresponding to the invalid invoices from the information matching module;
[0062] The invoice processing module invalidates or cancels the invoices in the invalid invoice subset library D module, and then reissues a new invoice based on the invoice information in the invoice information subset library E module.
[0063] Embodiment 1:
[0064] See also Figure 1 and Figure 2 , a method for identifying invoice cancellation and redemption, the method comprising the following steps:
[0065] S1: 100 invoices to be identified are input into the invoice scanning and identification module to establish an invoice information database A, which is stored in the invoice information database A module. The invoice is processed into information fragments and processed into 4 fields, namely, the purchaser information segment, the seller information segment, the tax total information segment, and the goods or taxable labor and service name information segment;
[0066] S2: Enter the invoicing information corresponding to all invoices that need to be identified and establish an invoicing information database B, which is stored in the invoicing information database B module;
[0067] S3: Put invoice a in invoice information database A into invoice information database B for matching. The corresponding information segment of invoice a is Company A, Company B, 100 yuan, and catering expenses.
[0068] S4: If invoice a does not get 100% matching degree in the invoice information database B, then invoice a is put into the invalid invoice subset library D, which is stored in the invalid invoice subset library D module, and the matching information of invoice a is recorded as follows: there is 1 invoice information with a matching degree of 75%, 5 invoice information with a matching degree of 50%, and 10 invoice information with a matching degree of 25%, and the invalid invoice subset library D is stored in the invalid invoice subset library D module;
[0069] S5: First, check the invoice information with a matching degree of 75% in descending order from 75% to 25%, and put this invoice information into the invoice information database A for matching. The invoice information with a matching degree of 100% is 0, the invoice information with a matching degree of 75% is 1, the invoice information with a matching degree of 50% is 4, and the invoice information with a matching degree of 25% is 8. The invoice information with a matching degree of 75% is invoice a. The invoice information with a matching degree of 75% found in S4 is the correct invoice information of invoice a, and the invoice information is recorded in the invoice information subset library E, and the invoice information subset library E is stored in the invoice information subset library E module;
[0070] S6: Identify the month of invoice a. The year and month on the invoice are January 2019, and the year and month on the computer system are January 2019, so invoice a is the invoice of that month;
[0071] S7: In the invoice processing module, invoice a finds the corresponding invoice information in the invoice information subset library E and performs cancellation and reopening processing.
[0072] Embodiment 2:
[0073] See also Figure 1 and Figure 2, a method for identifying invoice cancellation and redemption, the method comprising the following steps:
[0074] S1: 100 invoices to be identified are input into the invoice scanning and identification module to establish an invoice information database A, which is stored in the invoice information database A module. The invoice is processed into information fragments and processed into 4 fields, namely, the purchaser information segment, the seller information segment, the tax total information segment, and the goods or taxable labor and service name information segment;
[0075] S2: Enter the invoicing information corresponding to all invoices that need to be identified and establish an invoicing information database B, which is stored in the invoicing information database B module;
[0076] S3: Put invoice b in invoice information database A into invoice information database B for matching. The corresponding information fragment of invoice b is Company A, Company B, 100 yuan, accommodation fee;
[0077] S4: If invoice a is checked in the invoice information database B and has a 100% match, invoice b is put into the invoice valid subset library C, and the invoice valid subset library C is stored in the invoice valid subset library C module.
[0078] Example 3
[0079] See also Figure 1 and Figure 2 , a method for identifying invoice cancellation and redemption, the method comprising the following steps:
[0080] S1: 100 invoices to be identified are input into the invoice scanning and identification module to establish an invoice information database A, which is stored in the invoice information database A module. The invoice is processed into information fragments and processed into 4 fields, namely, the purchaser information segment, the seller information segment, the tax total information segment, and the goods or taxable labor and service name information segment;
[0081] S2: Enter the invoicing information corresponding to all invoices that need to be identified and establish an invoicing information database B, which is stored in the invoicing information database B module;
[0082] S3: Put invoice c in invoice information database A into invoice information database B for matching. The corresponding information fragment of invoice a is Company A, Company B, 100 yuan, and labor fee.
[0083] S4: If invoice c does not get 100% matching degree in the invoice information database B, then invoice c is put into the invalid invoice subset library D, which is stored in the invalid invoice subset library D module, and the matching information of invoice a is recorded as follows: there is 1 invoice information with a matching degree of 75%, 2 invoice information with a matching degree of 50%, and 5 invoice information with a matching degree of 25%, and the invalid invoice subset library D is stored in the invalid invoice subset library D module;
[0084] S5: In order of matching degree from 75% to 25% from high to low, first verify the invoice information with a matching degree of 75%, and put this invoice information into the invoice information database A for matching. The invoice information with a matching degree of 100% is 1, and the invoice c has been issued correctly with the invoice information with a matching degree of 75% in the invoice information database B; then put the two invoice information with a matching degree of 50% into the invoice information database A. One of the invoice information has an invoice information with a matching degree of 100% in the invoice information database A, and the other invoice information has a matching degree of 100% in the invoice information database B. There is no invoice information with 100% matching in the invoice information database A, but there is one invoice information with 75% matching, two invoice information with 50% matching, one of which is invoice c, and four invoice information with 25% matching. Then, the invoice information with 50% matching degree found in S4 and the invoice information with 50% matching degree with the invoice information database A is the correct invoice information of invoice a, and the invoice information is recorded in the invoice information subset library E, and the invoice information subset library E is stored in the invoice information subset library E module;
[0085] S6: Identify the month of invoice c. The year and month on the invoice are December 2018, and the year and month on the computer system are January 2019. Invoice a is not an invoice for this month.
[0086] S7: In the invoice processing module, invoice c finds the corresponding invoice information in the invoice information subset library E and performs redemption processing.
Claims
1. A method for identifying invoice cancellation and redemption, characterized by: The identification method comprises the following steps: S1: Enter all invoices that need to be identified and establish an invoice information database A. After the invoices are entered, the information is fragmented into N fields, where N is a positive integer greater than or equal to 1; S2: Enter the invoicing information corresponding to all invoices that need to be identified and establish an invoicing information database B; S3: put any invoice a in the invoice information database A into the invoice information database B for matching; S4: Check the matching degree between invoice a and invoice information database B. If there is a piece of invoice information that matches invoice a 100%, the invoice content of invoice a is correct, and then invoice a is put into valid invoice subset database C. If there is no piece of invoice information that matches invoice a 100%, the invoice content of invoice a is incorrect, and then invoice a is put into invalid invoice subset database D. S5: Put any invoice b in the invalid invoice subset library D into the invoice information data B again for matching, and obtain the matching degree. According to the matching degree from (N-1) / N to 1 / N from high to low, check the invoice information corresponding to each matching degree. The number of invoice information corresponding to each matching degree is recorded as m, where m is a natural number. Select m invoice information with a matching degree of (Nn) / N, and put the m invoices into the invoice information database A for matching. The matching degree is (Nn) / N. There are q invoice information, q is a positive integer, and one of the q invoice information is invoice b. Then the invoice information is the correct invoice information of invoice b. Then the correct invoice information corresponding to invoice b is recorded in the invoice information subset database E; Select m invoice information with a matching degree of (Nn) / N, and put the m invoices into the invoice information database A for matching. If the matching degree is 100%, select invoice information with matching degrees of (Nn-1) / N, (Nn-2) / N to 1 / N in turn and repeat the above steps until the correct invoice information corresponding to invoice b is found and recorded in the invoice information subset database E; S6: Identify the invoice month for the invoices in the invalid invoice subset library D. If the year and month on the invoice are consistent with the year and month on the computer system, the invoice is an invoice for the current month. If any of the year and month numbers on the invoice is less than the corresponding year and month number on the computer system, the invoice is not an invoice for this month. S7: Find the corresponding invoicing information in the invoicing information subset library E for the invoice of the current month obtained in step S6, and cancel it; and cancel the non-monthly invoice obtained in step S6.
2. A method for identifying invoice cancellation and redemption according to claim 1, characterized in that: The invoice in step S1 is processed into four fields, namely, the purchaser information field, the seller information field, the taxable total information field, and the goods or taxable labor or service name information field.
3. A method for identifying invoice cancellation and redemption according to claim 1, characterized in that: The invoicing information in step S2 includes sales contracts and purchase contracts.
4. A method for identifying invoice cancellation and redemption according to claim 1, characterized in that: The valid invoice subset library C and the invalid invoice subset library D in step S4 are subset libraries of the invoice information database A.
5. A method for identifying invoice cancellation and redemption according to claim 1, characterized in that : In step S5, n is greater than or equal to 1 and less than or equal to N-1, n is a positive integer, and the selection of the matching degree starts from (N-1) / N.
6. A method for identifying invoice cancellation and cancellation according to claim 1, characterized in that: In the step S2, the invoicing information entry content corresponds to the information segment of the invoice.
7. A recognition system for a method for recognizing a voided invoice or a redemption invoice according to any one of claims 1 to 6, characterized in that: The identification system includes an invoice scanning and identification module, an invoice information database block A, an invoicing information database module B, an information matching module, an invoice valid subset library module C, an invoice invalid subset library block D, an invoicing information subset library module E, and an invoice processing module; The invoice scanning and recognition module scans and recognizes the invoice through optical character recognition technology, and performs information fragmentation processing on the invoice according to needs; The invoice information database A module stores the invoice information obtained from the invoice scanning and recognition module; The invoicing information database module B classifies and stores the corresponding invoice information fragments of the sales contract and the purchase contract; The information matching module matches the information in the invoice information data module A with the information in the invoicing information database module B; The invoice valid subset library C module stores the valid invoice information obtained from the information matching module; The invalid invoice subset library D module stores the invalid invoice information obtained from the information matching module; The invoicing information subset library E module stores the invoicing information corresponding to the invalid invoices from the information matching module; The invoice processing module invalidates or cancels the invoices in the invalid invoice subset library D module, and then reissues a new invoice based on the invoice information in the invoice information subset library E module.
Citation Information
Patent Citations
Electronic invoice issuing method
CN106504040A
Blank invoice cancellation treatment method, storage medium and server
CN107784530A