JPWO2020012539A1 - Journal element analysis device, accounting system, journal element analysis method, journal element analysis program - Google Patents

Abstract

In the accounting processing device 10, the journal element analysis unit 11 verifies the journal elements related to the amount of money with respect to the extracted journal elements, or collates the journal element extraction results extracted by a plurality of journal element extraction AIs, and A warning display is added to the consistent journal entry elements.

Description

The present invention relates to a journal element analysis device, an accounting system, a journal element analysis method, and a journal element analysis program for extracting journal elements from voucher image data such as receipts and receipts.

Conventionally, as a voucher accounting process, tax accountants, accountants, and bookkeeping personnel read each item of information listed on the voucher and, for example, enter journal entry elements such as date, business partner, and amount in the ledger and respond to the journal entry elements. I entered the journals by empirically judging the account items I made.

As described above, it is not efficient for humans to read the vouchers one by one for journal entry, and the accuracy of the journals depends largely on the experience of the person in charge.Therefore, there is a problem that the accuracy of the journals varies. It was

Therefore, by using an OCR (Optical Character Reader) device to read the content of the voucher as electronic data and send it to the journal analysis center system via the Internet, the result of the journal entry for the bookkeeping transaction shown in the voucher. A so-called cloud-type accounting processing system that enables users to refer to has been developed (see Patent Documents 1 and 2).

More specifically, in the technique described in Patent Document 1, the voucher data captured by a mobile terminal or the like is analyzed by the journal element extraction means to extract journal element information as text format data. Then, for the registered user who owns the accounting software, the journal of the transaction described in the voucher and the similar transaction is extracted by referring to the past history of the registered user, and the recommended journal is presented according to the usage frequency. On the other hand, for non-registered users, the journal with the largest number of users of all users (many individuals and companies nationwide) is presented to the user as a recommended journal. Further, in Patent Document 1, when a new journal is generated or the journal is corrected on the user side, the learning means updates the database used for the journal so that the more journals are used by each user, the more the journal results are obtained. Optimized.

Further, in the technology described in Patent Document 2, transaction information is digitized by manual input or an OCR device, and inference from general commercial transactions (general inference) is performed, which is similar to a journal entered in the past. Journaling is performed by three inferences: inference (historical inference) and inference (clearing inference) from accounting events that occurred in advance such as accounts receivable and debt.

JP, 2014-235484, A JP, 2007-304643, A

In the above-mentioned Patent Documents 1 and 2, the journal element is extracted from the voucher by using the OCR device, but the display format of the voucher is various, and it is not easy to increase the accuracy of extracting the journal element. In particular, if the journal entry elements related to the product price, the total amount, the tax amount, the deposit amount, the change amount, etc. are incorrect, there is a high possibility that a large accounting problem will occur. Therefore, it is still necessary to manually check the extracted journal entry elements, and the work efficiency remains low.

The present invention has been made to solve such a problem, and an object of the present invention is to improve the efficiency of the work of inputting the journal entry described in the voucher and ensure the accuracy of extracting the journal entry. It is possible to provide a journal element analysis device, an accounting processing system, a journal element analysis method, and a journal element analysis program that can realize a more accurate automatic journal.

In order to achieve the above-mentioned object, the journal element analyzing apparatus according to the present invention is a journal element extracting unit capable of extracting journal elements from image data of a voucher, and the amount of journal elements extracted by the journal element extracting unit. And a consistency determining unit that performs a verification using a relational expression related to the journalizing element related to the journalizing element, a consistency determining unit that determines consistency of the journalizing element extracted by the journalizing element extracting unit from the verification result by the verification unit, and the consistency. An automatic correction unit that automatically corrects the inconsistent journal entry element by using the relational expression when the determination unit determines that there is a mismatch.

In the above-described journal element analysis device, the journal element extraction unit has a plurality of journal element extraction AIs for extracting journal elements from voucher image data, and image data of the same voucher is obtained by the plurality of journal element extraction AIs. Journal entry elements can be extracted from each of the journal entries, and a matching section that matches the journal entry elements extracted by the plurality of journal element extraction AIs is provided, and the consistency determination section determines the journal entry based on the result of the matching section. The consistency of the journal entry elements extracted by the element extraction unit may be determined.

Further, in the above-mentioned journal element analysis device, further, from the journal element extracted by the journal element extracting unit, a specific voucher for which the relational expression does not hold is determined, and for the journal element of the specific voucher, the verifying unit is used. A distribution unit may be provided that distributes the abutment by the abutment unit without performing the verification.

Further, in the above-mentioned journal element analysis device, further, display data of the journal elements extracted by the journal element extracting unit is generated, and the journal elements related to the determination result of the consistency determining unit are determined according to the consistency. A journal element output unit that gives different displays may be provided.

In order to achieve the above-mentioned object, the journal element analyzing apparatus according to the present invention is a journal element extracting unit capable of extracting journal elements from image data of a voucher, and the amount of journal elements extracted by the journal element extracting unit. A journalizing element that performs verification using a relational expression relating to the journalizing element, a consistency determining section that determines the consistency of the journalizing element extracted by the journalizing element extracting section from the verification result by the verifying section, and the journalizing element And a journalizing element output section for generating display data of the journalizing elements extracted by the extracting section, and displaying differently depending on the consistency of the journalizing elements related to the determination result of the consistency determining section. Good.

In order to achieve the above-mentioned object, the journal entry element analysis device according to the present invention has a plurality of journal entry element extraction AIs for extracting journal entry elements from image data of a voucher, and the same journal entry element extraction AI is used. A journal element extracting unit that can extract journal elements from the image data of the voucher, a matching section that matches the journal elements extracted by the plurality of journal element extracting AIs, and the journal element extraction from the matching result by the matching section. Consistency determination unit that determines the consistency of the journal entry elements extracted by the section, and when the consistency determination unit determines that there is a mismatch, the mismatch result is used to automatically correct the inconsistent journal element. And an automatic correction unit that does.

In order to achieve the above-mentioned object, the journal entry element analysis device according to the present invention has a plurality of journal entry element extraction AIs for extracting journal entry elements from image data of a voucher, and the same journal entry element extraction AI is used. A journal element extracting unit capable of extracting journal elements from image data of a voucher, a butt section for butting the journal elements having the same contents extracted by the plurality of journal element extracting AIs, and the butt result by the butt section. A consistency determination unit that determines the consistency of the journal element extracted by the journal element extraction unit and display data of the journal element extracted by the journal element extraction unit are generated, and are related to the determination result of the consistency determination unit. The journalizing element output unit that displays differently depending on the consistency of the journalizing element.

Further, in order to achieve the above-mentioned object, the accounting processing system according to the present invention includes a journal element analyzing apparatus, a journal element determining section for confirming the journal element, and a journal determined by the journal element confirming section. And a journalizing section that outputs account items based on elements.

In order to achieve the above-mentioned object, the journal element analysis method according to the present invention includes a journal element extracting step capable of extracting a journal element from image data of a voucher by a computer, and the journal element extracting step. A verification step of performing a verification using a relational expression relating to journal elements, a consistency determination step of determining the consistency of the journal elements extracted by the journal element extraction step from the verification result by the verification step, and the consistency determination If the process determines that there is a mismatch, an automatic correction process for automatically correcting the mismatched journal entry element using the relational expression is executed.

In order to achieve the above-mentioned object, the journal element analysis method according to the present invention includes a journal element extracting step capable of extracting a journal element from image data of a voucher by a computer, and the journal element extracting step. Of the journal elements, a verification step for performing a verification using a relational expression relating to the journal element related to the amount of money, and a consistency determination step for determining the consistency of the journal element extracted by the journal element extraction step from the verification result by the verification step. And a journal element outputting step of generating display data of the journal elements extracted by the journal element extracting step, and displaying differently depending on the consistency of the journal elements related to the determination result of the consistency determining step. , Run.

In order to achieve the above-mentioned object, the journal element analysis method according to the present invention uses a plurality of journal element extraction AIs for extracting journal elements from image data of a voucher by a computer, and the plurality of journal element analysis AIs. From the journalizing element extraction step of extracting journalizing elements from the image data of the same voucher by the extraction AI, the matching step of matching the journalizing elements extracted by the plurality of journalizing element extraction AIs, and the matching result of the matching step. A consistency determination step of determining the consistency of the journal entry elements extracted in the journal entry element extraction step, and an inconsistent journal entry using the match result when it is determined that there is inconsistency in the consistency determination step. And an automatic correction step of automatically correcting the element.

In order to achieve the above-mentioned object, the journal element analysis method according to the present invention uses a plurality of journal element extraction AIs for extracting journal elements from image data of a voucher by a computer, and the plurality of journal element analysis AIs. From the journalizing element extraction step of extracting journalizing elements from the image data of the same voucher by the extraction AI, the matching step of matching the journalizing elements extracted by the plurality of journalizing element extraction AIs, and the matching result of the matching step. A consistency determination step of determining the consistency of the journal entry elements extracted by the journal entry element extraction step, and display data of the journal entry elements extracted by the journal entry element extraction step are generated. A journal element outputting step of displaying a different display for the related journal elements depending on the consistency.

Further, in order to achieve the above-mentioned object, the journal element analysis program causes a computer to execute the above-mentioned journal element analysis method.

According to the present invention using the above means, it is possible to improve the efficiency of the input operation of the journal entry described in the voucher, ensure the accuracy of the extraction of the journal entry, and realize a more accurate automatic journal. it can.

It is a system configuration diagram showing an accounting processing system including a journalizing element analysis unit according to an embodiment of the present invention. It is the flowchart which showed the flow of the journal element analysis performed by the journal element analysis part. It is an example of a display displayed on a display part.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a system configuration diagram showing an accounting processing system including a journalizing element analysis unit according to an embodiment of the present invention, and the configuration of the present embodiment will be described based on the diagram.

As illustrated in FIG. 1, an accounting processing system 1 according to the present embodiment includes accounting processing on each device on the user side and accounting processing service provider side via a communication network 2 such as the Internet or a VPN (Virtual Private Network). It is configured to be connected to the device 10. Although only one user is shown in FIG. 1 for simplification of description, the accounting processing device 10 can be connected to a plurality of users via the communication network 2.

The user is, for example, an expert such as a tax accountant and an accountant, a corporation or an individual who directly performs accounting processing, and has at least the reading device 3 and the information terminal 4.

The reading device 3 is an optical device such as a scanner or a camera, and is a device that can capture a voucher as image data. In addition, the wording "voucher" in the present embodiment and the claims means receipts and receipts, other receipts, invoices, delivery notes, bankbooks of financial institutions, documents that are financial proofs for accounting, electronic It also includes transaction information using IC cards such as money.

The information terminal 4 is, for example, a personal computer (hereinafter referred to as a PC), a mobile terminal such as a smartphone, a tablet PC, and a mobile phone, and is a terminal capable of displaying at least web information.

The user can acquire the image data of the voucher with the reading device 3 and transmit it to the accounting processing device 10 with the information terminal 4, and can receive the information from the accounting processing device 10. Although the reading device 3 and the information terminal 4 are shown as separate bodies in FIG. 1, the reading device 3 and the information terminal 4 may be integrated as in a mobile terminal with a camera. In addition, the user does not need to own the reading device 3, and the image data of the voucher read by an external reading device may be acquired via e-mail or web, for example.

On the other hand, the accounting processing service provider (hereinafter, also simply referred to as a service provider) is a company that provides accounting processing service by so-called cloud computing, and manages the accounting processing device 10.

The accounting processing device 10 (accounting processing system) has one or more servers (computers) that execute journalizing processing based on a program, and functionally mainly extracts journal elements from image data of vouchers and analyzes them. A journal element analysis unit 11 (journal element analysis device) that performs the above, a display unit 12 that displays the analysis result of the journal elements, a journal element confirmation unit 13 that confirms the analyzed journal elements, and a journal element that has been confirmed. It has a journalizing unit 14 that performs automatic journaling, and a learning system 15 that generates journal element extraction AI and journal AI.

Further, the journal entry element analysis unit 11 includes a journal entry element extraction unit 20, a distribution unit 21, a verification unit 22, a matching unit 23, a consistency determination unit 24, an automatic correction unit 25, and a journal entry element output unit 26. ..

The journal element extracting unit 20 has a function of receiving image data of a voucher from a user and extracting journal elements from the image data. Specifically, the journal entry element extraction unit 20 has a plurality of journal entry element extraction AIs for extracting journal entry elements from the voucher image data, and the plurality of journal entry element extraction AIs respectively performs journal entry from the image data of the same voucher. Elements can be extracted. In this embodiment, the first journal entry element extraction AI and the second journal entry element extraction AI are used. The plurality of journal element extraction AIs have the same function of extracting journal elements from the voucher image data, but are AIs that have been subjected to different learning such as different learning data in the learning system 15.

The journal entry element extraction AI identifies a portion (position) corresponding to the journal entry element from the voucher image data, and extracts the journal entry element corresponding to the content of the identified portion as text. That is, the journal entry element extraction AI, in the learning system 15, specifies a region including a portion corresponding to the journal entry element from the image data by machine learning, and outputs the journal entry element corresponding to the content of the designated portion as text. It is AI who learned that. For example, the journal entry extraction AI specifies the date part, the amount part, the supplier part, and the summary part in the image data of the voucher, and recognizes the numbers in the date part and the amount part to recognize the date and the amount. Is output, the characters are recognized in the portion corresponding to the supplier and the portion corresponding to the abstract, and the text of the supplier and abstract is output.

Further, the journal entry element extraction unit 20 not only recognizes characters and numbers by the journal entry element extraction AI and outputs the text, but also, for example, when only the logo mark, the imprint or the telephone number is extracted, it is not shown. It may be output as the text of the supplier name by searching the company information DB that stores the company information in advance or by searching the information published on the Internet. If there is no description of the address such as a receipt, the user name who sent the image data may be output as the text of the transaction source, or the company that is the customer of the user may be set in advance. The image data transmitted from the user may output the set company name as the text of the transaction source.

Note that the journal entry element extraction AI is an image recognition AI that identifies a portion (position) corresponding to the journal entry element in the image data of the voucher, and a character recognition AI that extracts the journal entry element corresponding to the content of the identified portion as text. The two AIs may be used.

The journal entry elements include, for example, date, amount of money, business partner, summary (provided, including product name), business partner (including address), and corresponding numbers, characters, figures (for example, logo mark, imprint, other companies). There is an identifiable pattern) and the appearance of a voucher (for example, the size and color of a passbook or receipt).

For example, for a date, the journal entry element extraction unit 20 identifies the characters such as “date”, “year”, “month”, and “day” and the numeric parts before, after, and above and below symbols such as “/”. Regarding the amount of money, symbols such as "\" and product names, "amount", "deposit", "subtotal", "total", "tax", "discount", "yen", "payment", "deposit", "balance", etc. Identify the front and back and top and bottom numbers of the characters related to. Then, in the present embodiment, for the amount of money, a journal element is extracted for each character or number related to the amount of money, such as “total 1000 yen” and “change 50 yen”. Note that the journal entry extraction unit 20 can also recognize the meaning of the extracted characters, and can identify the accounting relationship such as, for example, “subtotal” is the sum of the prices of products.

In addition, regarding the business partners, specify the character parts before and after the characters such as "stock", "(shares)", "(ka)", the logo mark, the telephone number, and the appearance of the voucher, and based on this information. Identify the part corresponding to the company name or individual name. For the summary, specify the character portion that follows the character such as "But". For the transaction source, specify the part of the character that precedes the character such as "sama".

The journalizing element is not limited to this, and the numbers, characters, and figures used for extracting the journalizing element are not limited to this. For example, if the voucher contains the quantity of purchased items, the quantity may be included as a journal entry element. May be included as a journal entry element. In addition, a number (corporate number, establishment number) set to identify each company may be extracted.

The sorting unit 21 connected to the journal element extracting unit 20 determines a specific voucher for which the relational expression used in the verification unit 22 described later is not satisfied from the extracted journal elements, and determines the journal element of the specific voucher. Allocates so that the verification unit 22 does not perform verification and only the abutment unit 23 performs verification.

Specifically, the distribution unit 21 determines the type of voucher from the journal elements extracted by the journal element extraction unit 20, and sorts the voucher that can be verified and the voucher that cannot be verified. Specifically, if the relational expression used in the verification unit 22 has all the journalizing elements for which the relational expression is satisfied, it is a voucher that can be verified, and if the relational expression is not satisfied, it is a voucher that cannot be verified. Distribute. In addition, a relational expression used by the verification unit 22 in business practice or law such as a voucher in which only tax-included amount is described in advance such as a receipt for transportation such as taxi, train, handwritten receipt for restaurant, etc. The voucher that cannot be satisfied may be stored in advance in a storage unit (not shown) or the like, and the voucher determined from the extracted journal element may be sorted according to whether it is the registered specific voucher.

The checkout unit 22 has a function of performing a check using a relational expression relating to the journal entry element regarding the amount of money among the extracted journal entry elements. As a relational expression, for example, deposit = total + change, total (or subtotal) = total tax-included product amount, consumption tax amount = (total tax-included product amount-consumption tax amount) x consumption tax rate, tax-excluded amount = tax-included amount / There is consumption tax, etc., and it may be calculated from a simultaneous equation consisting of a plurality of relational expressions. As a relational expression, in addition to this, when the consumption tax rate differs from 8% to 10% depending on the product, the consumption tax amount=8% target product total amount×0.08+10% product total amount×0.10, You may use the relational expression. The relational expression based on such accounting is stored in a storage unit (not shown), and the verification unit 22 calls and uses an appropriate relational expression from the storage unit according to the journalizing element regarding the extracted amount.

In the verification in the verification unit 22, for example, it is determined from the extracted numerical values of the journalizing elements whether the relational expression is satisfied (whether the equality is satisfied), and whether or not the relationship is satisfied is output as a verification result. In particular, when the relational expression does not hold, information such as which element of which expression does not hold is also included in the verification result. The check unit 22 may check each journal element related to the amount of money extracted by the plurality of journal element extraction AIs, or may set one journal element extraction AI to be used for the check in advance. , Journal elements with high reliability of extraction may be used. The reliability here is the accuracy determined by the journal element extraction AI, and can be expressed as a percentage, for example.

The matching unit 23 matches the journal elements of the same type extracted by the journal element extraction AI in the journal element extraction unit 20. For example, the total amount of money extracted by the first journal entry element extraction AI and the total amount of money extracted by the second journal entry element extraction AI are matched. Then, for each journal entry element, the journal entry elements that match each other and the journal element that does not match each other are output as the matching result. Note that, in the present embodiment, the reconciling unit 23 is described as reconciling the journalizing element related to the amount of money among the extracted journalizing elements, like the checking unit 22, but the reconciling unit 23 performing the reconciling is the monetary accounting element. It is not limited to the journal entry element related to.

The consistency determination unit 24 determines the consistency of the extracted journal entry elements from the verification result by the verification unit 22 and the matching result by the matching unit 23. In addition, for a voucher that cannot be verified, the consistency of the journal entry elements is determined based only on the result of the match.

Specifically, the consistency determination unit 24 determines, for example, that all the journal elements are consistent from the verification result and the match result, that there is an inconsistent journal element, but that the journal element can be corrected, and the journal element that cannot be corrected. It is possible to output a determination result such as "there is".

The automatic correction unit 25 has a function of automatically correcting the inconsistent journal entry elements determined by the consistency determination unit 24 using the verification result or the match result. For example, when there is a journal entry element that prevents the simultaneous equations of the relational expressions used in the verification unit 22 from being satisfied, the automatic correction unit 25 changes the numerical value to a value with which the simultaneous equation is established for the journal element. Alternatively, when there is a mismatching journal entry element from the matching result, the automatic correction unit 25 adopts the journal entry element extracted by the journal element extraction AI with high reliability and changes the inconsistent portion.

The journal entry element output unit 26 has a function of outputting the journal entry elements extracted by the journal entry element extraction unit 20, including the automatically corrected journal entry elements, as display data that can be displayed on the display unit 12. In particular, the journal entry element output unit 26 gives a different display depending on the consistency determined by the consistency determination unit 24. For example, in the present embodiment, the journal entry element output unit 26 normally displays when there is no inconsistent journal entry element, and the first warning display (red flag display) is displayed for the inconsistent journal entry element that cannot be automatically corrected. A second warning display (yellow flag display) is attached to the automatically corrected inconsistent journal entry elements.

The warning display may be any display that is different from the normal display, and the display format is not particularly limited.For example, the text of the inconsistent journal entry element is underlined, the text color is different, or a flag is displayed near the text. Express by doing. Also, a plurality of types of warning display may be set according to the consistency, or only one type may be set.

The display unit 12 is, for example, a display of the accounting processing device 10, and has a function of displaying display data indicating the extraction result of the journal element output from the journal element output unit 26. The display data generated by the journal element output unit 26 is transmitted not only to the display unit 12 of the accounting processing device 10 but also to the information terminal 4 of the user so that the user can check the extraction result of the journal element. Good.

Next, the journal entry element confirming unit 13 carries out a confirming process for confirming the journal element displayed on the display unit 12 by the service provider or the program for confirming the journal element. For example, the person in charge on the service provider side confirms the journal entry elements displayed on the display unit 12, performs the confirmation operation as is for the journal entry elements having no problem, and corrects the journal entry elements having the problem. Then, perform the confirmation operation. If a warning display is attached to the inconsistent journal element or the automatically corrected journal element in the journal element output unit 26, it becomes easy to find the corrected portion.

The journalizing unit 14 has a function of outputting an account item corresponding to the journalizing element determined by the journalizing element determining unit 13. The output of the account item according to the journal element is performed by, for example, journal AI. The journal entry AI is an AI of an automatic journal that the learning system 15 has learned in advance to output an account item for a journal element by machine learning. The account items output by the journalizing unit 14 are transmitted to the user's information terminal 4 as the journalizing result together with the journalizing elements.

The learning system 15 has a function of learning the journal element extraction AI and the journal AI described above, and supplying the learned AI. Specifically, the learning system 15 generates the journal element extraction AI by performing machine learning (so-called deep learning) on the basis of the learning data including the voucher image data and the journal elements included in the image data. Further, the learning system 15 generates a journal AI by performing machine learning on the basis of learning data including a journal element and an account item corresponding to the journal element.

In the accounting processing device 10 configured as described above, the journal element analysis unit 11 performs a check on the journal elements related to the amount of money with respect to the extracted journal elements, or a journal entry of the same type extracted by a plurality of journal element extraction AIs. Match the element extraction results and display a warning display for inconsistent journal elements.

Referring to FIG. 2, there is shown a flowchart showing a flow of the journal element analysis performed by the journal element analyzing section 11. The journal element analyzing method will be described in detail below with reference to the flowchart. The journal element analysis starts when the image data of the voucher from the user is received.

First, as step S1, the journal element analysis unit 11 extracts journal elements from the image data of the same voucher by the journal element extracting AI of the journal element extracting unit 20 (journal element extracting step).

In step S2, the distribution unit 21 establishes a relational expression that is not a specific voucher registered in advance from the extracted journal entry elements and that can be calculated from the journal entry elements related to the amount of money in the extracted journal entry elements. It is determined whether or not. If the determination result is true (Yes), the process proceeds to the next step S3. On the other hand, when the determination result is false (No), that is, when the voucher is a specific voucher such as a transportation receipt or a handwritten receipt, or when a relational expression that can be verified is not established from the extracted journal entry elements. Skips step S3 and proceeds to step S4.

In step S3, the verification unit 22 generates a relational expression based on accounting for the journal entry element related to the amount of money, and performs a verification using the relational expression (verification process).

In step S4, the matching section 23 matches the same type of journal elements related to the amount of money extracted by the plurality of journal element extraction AIs (matching step).

In step S5, the consistency determination unit 24 determines whether there is an inconsistent journal element from the verification result and the matching result in step S4 when the verification is performed in step S3 (consistency determination). Process). If the determination result is false (No), that is, if both the verification result and the match result match, the process proceeds to step S6.

In step S6, the journalizing element output unit 26 generates the display data of the journalizing element extracted in the normal display, displays it on the display unit 12, and ends the routine.

On the other hand, if the determination result of step S5 is false (No), that is, if there is an inconsistent journal entry element, the process proceeds to step S7.

In step S7, the consistency determination unit 24 determines whether or not the inconsistent journal entry element can be identified from the verification result or the match result (consistency determination step). If the determination result is false (No), that is, it is possible to recognize that there is an inconsistent journal entry element, but the journal entry element cannot be specifically specified, the process proceeds to step S8.

In step S8, the journal element output unit 26 attaches the first warning display (for example, red display) to the inconsistent journal element (in this case, the amount), and displays the display data including other extracted journal elements. It is generated and displayed on the display unit 12, and the routine ends.

On the other hand, if the determination result of step S7 is true (Yes), that is, if the inconsistent journal entry element can be specified, the process proceeds to step S9.

In step S9, the automatic correction unit 25 automatically corrects the inconsistent journal entry elements to correct values by using the verification result or the matching result.

Then, in the subsequent step S10, the journal element output unit 26 attaches the second warning display (for example, yellow display) to the inconsistent journal element (in this case, the amount), and also includes other extracted journal elements. The display data is generated, displayed on the display unit 12, and the routine ends.

Referring now to FIG. 3, a display example displayed on the display unit 12 is shown, and based on the drawing, a specific display of the journalizing elements will be described.

The display example of FIG. 3 is a display example when the journalizing elements are extracted from the image data of the receipt. The original image of the scanned receipt is displayed on the left side of the screen, and the journals corrected by the automatic correction unit 25 are displayed on the right side of the screen. The journal entry elements including the elements extracted by the journal entry element extraction unit 20 are displayed. That is, FIG. 3 is a display example when the processing of steps S9 and S10 of FIG. 2 is performed.

Specifically, the receipt is displayed with one transaction described for each image. In such a case, as indicated by the dotted frame in the original image, the journal entry element extraction AI identifies the portion 40a corresponding to the date, the portions 41a1 to 41a5 corresponding to the amount of money, and the portion 42a corresponding to the company name. The enlarged views 40b, 41b, 42b are displayed in the read result. The portion corresponding to the amount of money consists of a product price 41a1 consisting of “coffee ¥540 and tea ¥540”, a total amount of money 41a2 consisting of “total ¥1,080”, and “(consumption tax 8% ¥80)”. A consumption tax amount 41a3, a deposit amount 41a4 composed of "deposit \1,100", and a change amount 41a5 composed of "change \20" are designated, respectively. The dotted line may not be actually displayed on the display unit 12.

In the read result, the date, the amount of money, and the company name, which are made into text by the journal entry element extraction AI, are described on the corresponding enlarged view.

Specifically, with respect to the portion 40a corresponding to the date and the portion 42a corresponding to the company name, the journal element extraction AI correctly extracts the journal elements in the reading result.

On the other hand, for the portions 41a1 to 41a5 corresponding to the amount of money, the “8” portion of the total amount of money 41a2 is faint, so the first journal entry element extraction AI erroneously recognizes it as “1,000 yen” in the reading result. However, in the read result, the numerical value automatically corrected to “1080 yen” is displayed, and the second warning display 44 is provided next to it. During the normal display, the flag like the second warning display 44 of FIG. 3 is not displayed, or the flag of the color indicating the normal display is displayed. On the other hand, if there is a mismatch that cannot be automatically corrected, a first warning display such as a color difference is displayed instead of the second warning display 44.

Here, the method of automatic correction will be specifically described. In the automatic correction based on the verification result in FIG. 3, as shown in the verification result in FIG. 3, two relational expressions of “deposit=total+change, total=commodity price including tax−consumption tax” are used. When the extracted amounts are put into these relational expressions, a simultaneous equation consisting of the following three expressions is obtained.
(1) Custody (1100) = Total (1000) + Change (20)
(2) Total (1000) = tax-included product price (540 + 540)
(3) Consumption tax amount (80)={Product price including tax (540+540)−Consumption tax amount (80)}×Consumption tax rate (8%)

However, it is possible to determine that there is a mismatch because the equal signs in the expressions (1) and (2) do not hold, and in this case, it is possible to specify from calculation that the “total (1000)” is a mismatch. Furthermore, the sum is derived as a verification result that "1080" is a matching numerical value, and this numerical value is reflected as a reading result.

Further, as shown in the match result of FIG. 3, the “total” extracted by the first journal entry element extraction AI is “¥1,000”, while it is extracted by the second journal entry element extraction AI. If the “total” is “¥1,080”, it can be determined as a result of the inconsistency that the total amount 41a2 is inconsistent due to these inconsistencies. For example, the reliability of the second journal entry element extraction AI If the reliability is higher than the reliability of the first journal entry element extraction AI, the extraction result of the second journal entry element extraction AI can be automatically corrected to “¥1,080”.

As described above, in the accounting processing system 1 according to the present embodiment, the journal entry element analysis unit 11 performs the verification using the relational expression relating to the journal entry element regarding the amount, and uses the relational expression for the inconsistent journal entry element. Automatic correction is possible. As a result, it is possible to omit manual confirmation work and correction work of journal elements, reduce the work of inputting journal elements, and improve work efficiency. In addition, it is possible to reduce the input of inconsistent journal entry elements by automatic correction, and it is possible to ensure the accuracy of journal entry element extraction.

Further, the journal entry element analysis unit 11 extracts journal elements from the image data of the same voucher by a plurality of journal element extraction AIs, compares the journal elements with each other to determine the consistency, and matches the inconsistent journal elements. Automatic correction is possible using the results. This also makes it possible to omit the manual confirmation work and correction work of the journal entry elements, reduce the correction work in the input work of the journal entry elements, and improve the work efficiency. Further, it is possible to reduce the input of inconsistent journal entry elements by the automatic correction, and it is possible to ensure the accuracy of the journal entry element extraction.

Further, the journal entry element analysis unit 11 discriminates a specific voucher for which the relational expression does not hold from the extracted journal entry elements, and does not perform the verification by the verification unit 22 for the journal element of the specific voucher, but only by the matching by the matching unit 23. By allocating so as to perform, it is possible to omit useless verification and improve processing efficiency.

Further, by displaying differently the journalizing elements related to the determination result of the consistency determining unit 24 according to the consistency, it is possible to easily grasp the inconsistent portion and the degree of inconsistency, and to further improve the work efficiency. It is possible to improve.

Then, the journalizing element is determined based on the result output by the journalizing element output unit 26, and the journalizing unit 14 performs the automatic journaling, whereby a more accurate automatic journaling can be realized.

From the above, the accounting device 10 including the journal element analysis unit 11 according to the present embodiment improves the efficiency of the journal element input work described in the voucher, and secures the accuracy of journal element extraction. It is possible to realize more accurate automatic journal entry.

Although the description of the embodiment of the present invention has been completed, the aspect of the present invention is not limited to this embodiment.

For example, in the above embodiment, the extraction of the journal entry element of the receipt is described as the voucher based on the display example of FIG. 3, but the type of voucher is not limited to this.

Further, in the above embodiment, the journalizing unit 14 performs the automatic journaling using the journal AI, but the journaling may be performed by an automatic journaling program that does not use the AI.

Further, in the above-described embodiment, the journal elements of the same type extracted by the plurality of journal element extraction AIs are abutted with each other regardless of whether the verification is performed or not, but only one of the verification and the abatement is performed. You may

1 accounting system 2 communication network 3 reader 4 information terminal 10 accounting processor 11 journal element analysis unit (journal element analyzer)
12 display section 13 journal element determining section 14 journal section 15 learning system 20 journal element extracting section 21 sorting section 22 verification section 23 match section 24 consistency determination section 25 automatic correction section 26 journal element output section

Claims (13)

A journal element extracting unit capable of extracting journal elements from the voucher image data,
Of the journal entry elements extracted by the journal entry element extraction unit, a verification unit for performing verification using a relational expression relating to the journal entry element related to the amount of money,
A consistency determination unit that determines the consistency of the journal element extracted by the journal element extraction unit from the verification result by the verification unit,
When it is determined that there is a mismatch by the consistency determination unit, an automatic correction unit that automatically corrects the inconsistent journal entry element using the relational expression,
A journal element analyzing device. The journal element extraction unit has a plurality of journal element extraction AIs for extracting journal elements from the voucher image data, and can extract journal elements from the same voucher image data by the plurality of journal element extraction AIs. Yes,
Furthermore, a butt section for butting the journal entry elements extracted by the plurality of journal entry element extraction AIs is provided,
The journal entry element analysis device according to claim 1, wherein the consistency determination unit determines the consistency of the journal entry elements extracted by the journal entry element extraction unit from the matching result by the matching section. Furthermore, a specific voucher in which the relational expression does not hold is determined from the journal entry elements extracted by the journal entry element extraction unit, and the journal element of the specific voucher is not verified by the verification unit and the match by the matching unit is performed. The journal element analysis device according to claim 2, further comprising a distribution unit that distributes the data to be executed. Further, there is provided a journal element output unit for generating display data of the journal elements extracted by the journal element extracting unit, and for the journal elements related to the judgment result of the consistency judging unit, displaying differently depending on the consistency. The journal element analysis device according to any one of claims 1 to 3, further comprising: A journal element extracting unit capable of extracting journal elements from the voucher image data,
Of the journal entry elements extracted by the journal entry element extraction unit, a verification unit for performing verification using a relational expression relating to the journal entry element related to the amount of money,
A consistency determination unit that determines the consistency of the journal element extracted by the journal element extraction unit from the verification result by the verification unit,
A journal element output unit for generating display data of the journal elements extracted by the journal element extracting unit, and for the journal elements related to the determination result of the consistency determining unit, a different display according to the consistency,
A journal element analyzing device. A journal element extraction unit that has a plurality of journal element extraction AIs for extracting journal elements from the voucher image data, and that can extract journal elements from the same voucher image data by the plurality of journal element extraction AIs.
An abutting section for abutting the journalizing elements extracted by the plurality of journalizing element extracting AIs;
A consistency determination unit that determines the consistency of the journalizing elements extracted by the journalizing element extracting unit from the result of the matching by the matching unit,
When it is determined that there is a mismatch by the consistency determination unit, an automatic correction unit that automatically corrects the inconsistent journal entry element using the match result,
A journal element analyzing device. A journal element extraction unit that has a plurality of journal element extraction AIs for extracting journal elements from the voucher image data, and that can extract journal elements from the same voucher image data by the plurality of journal element extraction AIs.
An abutting section for abutting the journal elements having the same contents extracted by the plurality of journal element extracting AIs;
A consistency determination unit that determines the consistency of the journalizing elements extracted by the journalizing element extracting unit from the result of the matching by the matching unit,
A journal element output unit for generating display data of the journal elements extracted by the journal element extracting unit, and for the journal elements related to the determination result of the consistency determining unit, a different display according to the consistency,
A journal element analyzing device. A journal element analyzing device according to any one of claims 1 to 7,
A journal element determining section for confirming the journal element,
A journalizing unit that outputs an account item based on the journalizing element determined by the journalizing element determining unit. By computer
A journal element extracting step capable of extracting journal elements from the voucher image data,
A check process for performing a check using a relational expression relating to the journal elements extracted in the journal element extracting step,
A consistency determination step of determining the consistency of the journal element extracted by the journal element extraction step from the verification result by the verification step,
If it is determined that there is inconsistency by the consistency determination step, an automatic correction step of automatically correcting inconsistent journal entry elements using the relational expression,
Journal element analysis method for executing. By computer
A journal element extracting step capable of extracting journal elements from the voucher image data,
Of the journal entry elements extracted by the journal entry element extraction step, a verification step for performing verification using a relational expression relating to the journal element related to the amount,
A consistency determination step of determining the consistency of the journal element extracted by the journal element extraction step from the verification result by the verification step,
A journal element outputting step of generating display data of the journal elements extracted by the journal element extracting step, and displaying differently depending on the consistency of the journal elements related to the determination result of the consistency determining step,
Journal element analysis method for executing. By computer
A journal element extracting step of using a plurality of journal element extracting AIs for extracting journal elements from the voucher image data, and extracting journal elements from the same voucher image data by the plurality of journal element extracting AIs, respectively.
A matching step of matching the journal elements extracted by the plurality of journal element extraction AIs,
A consistency determination step of determining the consistency of the journal entry elements extracted by the journal entry element extraction step from the result of the match by the match step;
When it is determined that there is a mismatch by the consistency determination step, an automatic correction step for automatically correcting the inconsistent journal entry element using the match result,
Journal element analysis method for executing. By computer
A journal element extracting step of using a plurality of journal element extracting AIs for extracting journal elements from the voucher image data, and extracting journal elements from the same voucher image data by the plurality of journal element extracting AIs, respectively.
A matching step of matching the journal elements extracted by the plurality of journal element extraction AIs,
A consistency determination step of determining the consistency of the journal entry elements extracted by the journal entry element extraction step from the result of the match by the match step;
A journal element outputting step of generating display data of the journal elements extracted by the journal element extracting step, and displaying differently depending on the consistency of the journal elements related to the determination result of the consistency determining step,
Journal element analysis method for executing. A journal element analysis program for causing a computer to execute the journal element analysis method according to any one of claims 9 to 12.

Images