WO2020179055A1

WO2020179055A1 - Voucher analysis device, accounting system, voucher analysis method, and voucher analysis program

Info

Publication number: WO2020179055A1
Application number: PCT/JP2019/009077
Authority: WO
Inventors: 森啓太郎; 小嶋勇志
Original assignee: ファーストアカウンティング株式会社
Priority date: 2019-03-07
Filing date: 2019-03-07
Publication date: 2020-09-10
Also published as: JPWO2020179055A1; JP6646308B1

Abstract

A voucher analysis unit 11 of an accounting system 1 is provided with: a region identification unit 21 which identifies an imprint region from image data of a voucher, and identifies a region that overlaps the imprint region and that corresponds to a journal entry element, such as an issuer region including an issuer element; a journal entry element extraction unit 22 which extracts journal entry elements including the issuer element from the issuer region identified by the region identification unit 21; a client database 23 which stores client information; and an issuer estimation unit 24 which compares the issuer element extracted by the journal entry element extraction unit 22 with the client information stored in the client database 23 to estimate the issuer of the voucher.

Description

Voucher analysis device, accounting system, voucher analysis method, voucher analysis program

The present invention relates to a voucher analysis device, an accounting processing system, a voucher analysis method, and a voucher analysis program that can estimate the issuer of a voucher from image data of a voucher such as an invoice or a receipt.

Conventionally, tax accountants, accountants, and bookkeepers read the information described in the voucher one by one as accounting processing of vouchers, and enter journal entry elements such as date, business partner, amount, etc. in the ledger and respond to the journal entry elements. The journal entry was made by empirically judging the account that was created.

In this way, if a human reads the voucher one by one and makes a journal entry, the work efficiency is poor, and the accuracy of the journal entry largely depends on the experience of the person in charge, and there is a problem that the accuracy of the journal entry varies. It was

Therefore, by using an OCR (Optical Character Reader) device, the contents of the voucher can be read as electronic data and transmitted to the journal analysis center system via the Internet. A so-called cloud-type accounting system has been developed that allows users to refer to (see Patent Document 1).

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 extracting means and the journal element information is extracted as text format data. Specifically, the character information group recorded in the image data of the voucher such as a receipt is recognized and converted into text data composed of items such as a user ID, a date, a purchaser name, and a purchased product name. .. Then, a product group corresponding to the product name included in the journal element information converted into text data is obtained from the product master, and a recommended journal is indicated using the mapping table corresponding to the product group (paragraph of Patent Document 1). See 0118 to paragraph 0128).

JP, 2014-235484, A

Although the journal element is extracted from the voucher using the OCR device in the above-mentioned Patent Document 1, the display format of the voucher is various, and the information described in the voucher is also various, so the accuracy of extracting the journal element is high. It's not easy to do. For example, when a plurality of company names are described in one voucher, it may not be possible to determine which company name is the issuer of the voucher.

On the other hand, invoices and receipts often have a company seal or the like stamped on the part where the issuer information is written. In addition, foreign matter may be mixed in when reading the voucher, the voucher itself may be scratched, the voucher may be a copy type and the characters may be faint, or the voucher may contain handwritten characters. In such a case, when the journal element is extracted from the voucher using the OCR device, the information of the issuer is partially lost and extracted, and the issuer cannot be accurately recognized. If the publisher cannot be accurately recognized in this way, there is a problem that humans must eventually confirm the voucher and work efficiency cannot be improved.

The present invention has been made to solve such a problem, and an object of the present invention is to be able to recognize the voucher issuer more quickly and accurately from the voucher image data, and the voucher issuer. It is intended to provide a voucher analysis device, an accounting processing system, a voucher analysis method, and a voucher analysis program capable of improving the efficiency of the input work.

In order to achieve the above-mentioned object, the voucher analysis apparatus according to the present invention uses an area identification unit for identifying a journal element area including a journal element from image data of the voucher, and a journal element area identified by the area identification unit. An issuer element extraction unit that extracts an issuer element indicating the issuer of the trademark, a storage unit that stores supplier information, an issuer element extracted by the issuer element extraction unit, and a storage unit that stores the supplier information. An issuer estimation unit that estimates the issuer of the voucher by collating with the obtained supplier information.

In the voucher analysis device described above, the issuer estimation unit issues the voucher according to the degree of agreement between the issuer element extracted by the issuer element extraction unit and the business partner information stored in the storage unit. The origin may be estimated.

Further, in the above-described voucher analysis device, the area identification unit identifies an area of a seal imprint from the image data of the voucher, identifies an issuer area that includes an issuer element and overlaps the area of the imprint, and issues the issue. The source element extraction unit may extract the issuer element from the issuer area identified by the area identification unit.

Further, in the above-described voucher analysis apparatus, the area identification unit can identify, as the journal entry element area, a FAX information area in which outgoing information is described when FAX is received from image data of the voucher, and the issuer element extraction is performed. The unit can extract the issuer element from the FAX information area identified by the area identification unit, and the issuer estimation unit includes the issuer element extracted from the FAX information area in the storage unit. You may collate with the stored supplier information.

Further, in the voucher analysis device described above, the area identification unit can identify an account information area in which bank account information is described as the journal element area, and the issuer element extraction unit can be used by the area identification unit. The issuer element can be extracted from the identified account information area, and the issuer estimation unit includes the issuer element extracted from the account information area and is associated with the business partner information stored in the storage unit. You may collate.

Further, in the voucher analysis device described above, the area identification unit can identify the number information area in which the unique number information peculiar to the business partner is described as the journal element area, and the issuer element extraction unit can identify the number information area. The issuer element can be extracted from the unique number information area identified by the area identification unit, and the issuer estimation unit stores the issuer element including the issuer element extracted from the unique number information area in the storage unit. You may collate with supplier information.

Further, in the voucher analysis device described above, the area identification unit can identify the e-mail address information area in which the e-mail address information is described as the journal element area, and the issuer element extraction unit can identify the area identification unit. The issuer element can be extracted from the e-mail address information area identified by, and the issuer estimation unit includes the issuer element extracted from the e-mail address information area and is stored in the storage unit. You may collate with information.

Further, in order to achieve the above object, the accounting system according to the present invention, the voucher analysis device described above, a journalizing unit that outputs an account item corresponding to the journalizing element of the voucher including the issuer element, Equipped with.

In order to achieve the above-mentioned object, the voucher analysis method according to the present invention is such that a computer identifies an area of a seal imprint from image data of the voucher, and an issuer including an issuer element that overlaps the area of the imprint. An area identification step for identifying an area, an issuer element extraction step for extracting the issuer element from the issuer area identified by the area identification step, and an issuer element extracted by the issuer element extraction step. The issuer estimation step of estimating the issuer of the voucher by collating with the business partner information stored in the storage unit is executed.

Further, in order to achieve the above-mentioned purpose, in the voucher analysis program, the computer is made to execute the above-mentioned voucher analysis method.

According to the present invention using the above means, the issuer of the voucher can be recognized more quickly and accurately from the image data of the voucher, and the efficiency of the input work of the issuer of the voucher can be improved.

It is a system block diagram which showed the accounting processing system including the voucher analysis part which concerns on one Embodiment of this invention. It is a flowchart which showed the flow of the voucher analysis executed by the voucher analysis unit. It is explanatory drawing which estimates the issuer from the invoice including the imprint.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

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

As shown in FIG. 1, the accounting processing system 1 according to the present embodiment accounts for each device on the user side and accounting processing on the accounting processing service provider side via a communication network 2 such as the Internet and 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, or 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 proof of financial transfer in accounting, electronic Transaction information using IC cards such as money shall also be included.

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 by the reading device 3 and transmit it to the accounting processing device 10 by the information terminal 4, and can also 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. Further, the user does not have to own the reading device 3, and for example, the image data of the voucher read by an external reading device may be acquired via an e-mail or a web.

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

The accounting processing device 10 has one or more servers (computers) that execute journalizing processing based on a program, and functionally mainly extracts journal entry elements including a voucher issuer element from voucher image data. Based on the confirmed journalizing element, the voucher analyzing section 11 (certificate analyzing device) for performing the analysis according to the above, the display section 12 for displaying the analysis result of the journalizing element, the journalizing element confirming section 13 for confirming the analyzed journalizing element It has a journal unit 14 that performs automatic journalizing, and a learning system 15 that generates AI such as an area identification AI and a journal AI.

Further, the voucher analysis unit 11 has an image acquisition unit 20, an area identification unit 21, a journal element extraction unit 22 (issuer element extraction unit), a supplier database 23 (storage unit), and an issuer estimation unit 24. ing.

The image acquisition unit 20 has a function of receiving the image data of the voucher read by the reading device 3 of the user. Although not shown, the image acquisition unit 20 may have a storage means for storing received image data.

The area identification unit 21 has an image analysis function of identifying the journal element area corresponding to the journal element from the image data of the voucher acquired by the image acquisition unit 20.

Journal elements include, for example, date, amount, business partner, description (including proviso and product name), business source (including address), and the corresponding numbers, letters, graphics (for example, logo mark, imprint, and other companies). There is a identifiable pattern) and the appearance of the voucher (eg, the size and color of a passbook or receipt).

For example, for a date, the area identification unit 21 can specify the numbers before and after the characters such as "date", "year", "month", "day", and the symbols such as "/". Regarding the amount of money, it is possible to specify the symbols such as “¥” and the numerical parts before and after or above and below the characters such as “amount of money”, “payment”, “deposit”, “balance” and “yen”. For business partners, specify the character parts before and after the characters such as "Co., Ltd.", "Co., Ltd.", "(F)", logo mark, telephone number, FAX number, and voucher appearance, and these It is possible to identify the part corresponding to the company name or personal name based on the information. As for the abstract, it is possible to specify the character part following the character such as "however". For the trading source, the part of the character before the character such as "sama" can be specified.

Note that the journal elements are not limited to this, and the numbers, letters, and figures used to extract the journal elements 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 factor, and if information such as the names and number of attendees is stated, the attendees and the number of attendees. May be included as a journal entry element. In addition, a unique number (corporate number, business office number) set to identify each company may be extracted. Other unique numbers of business partners that can identify business partners in this way include transaction codes, billing numbers, quotation numbers, order numbers, and the like.

Specifically, when the input image data is a scan of a passbook, the area identification unit 21 specifies an area including a part corresponding to a date, transaction details, payment, deposit, balance, and the like. The part corresponding to the journal entry element is specified. If the image data is a scan of a receipt or receipt, specify the area including the date part, logo or company name part, amount part, and proviso part to identify the part corresponding to the journal element. To do.

The area identification unit 21 identifies such a journal element area by the area identification AI. In the learning system 15, the area identification AI uses teacher data (learning data) consisting of image data of various vouchers and data indicating each journal element area in the image data, and uses machine learning in the image data. It is an AI learned to identify an area including a part corresponding to a journal element from.

In particular, the area identification unit 21 in the present embodiment identifies the area of the imprint in the image data of the voucher having the imprint indicating the issuer such as an invoice or a receipt, and further identifies the area of the imprint in the image data. It overlaps with the area and has a function to identify the issuer area including the issuer element.

The identification of such a seal imprint area and the issuer area may be performed by image analysis other than AI, or may be performed by area identification AI. In the case of the area identification AI, in the learning system 15, the teacher data (learning data) consisting of the image data of the voucher including the imprint and the data indicating the imprint area and the issuer area in the image data is used. The area identification AI is generated by learning to identify the seal imprint area and the issuing area from the image data by machine learning.

Note that the imprint is mainly a company seal for a corporation, a shop name seal for a sole proprietor, etc., and may be a circle seal in addition to a square seal. In addition, the issuer element corresponds to the business partner in the above journal element, and specifically, the company name, personal name (store name), address, telephone number, fax number, e-mail address, department name, person in charge, Etc. information is included.

Further, the area identification unit 21 in the present embodiment can identify the fax information area in which the starting information such as the fax number of the sender is described at the time of receiving the fax from the image data of the voucher. The FAX information area is mainly printed on the end of the voucher (for example, header or footer), and the area identification unit 21 may identify the area by area identification AI or image analysis, and regardless of these, the end of the voucher. The part area may be set as the FAX information area for identification.

The journal element extraction unit 22 has a character recognition function of outputting the journal elements in the area corresponding to each journal element identified by the area identification unit 21 as text. For example, the journal element extraction unit 22 recognizes numbers in the date part and the amount part, outputs the text of the date and the amount, and recognizes the characters in the part corresponding to the business partner and the part corresponding to the description. To output the text of the supplier or summary. The character recognition function may use the conventional OCR technology, or in the learning system 15, the teacher data (learning data) including the journal element portion in the image data and the text corresponding to the journal element is used. Alternatively, a text-based AI learned to output the text of the journal element by machine learning may be used.

In particular, the journal element extraction unit 22 in the present embodiment refers to the image data of the voucher having the imprint indicating the issuer, such as an invoice or a receipt, from the issuer area identified by the area identification unit 21. Extract elements. That is, the journal entry element extraction unit 22 extracts, as an issuer element, a text in the issuer area that overlaps with the seal imprint area.

Note that the journal entry element extraction unit 22 extracts the issuer element while excluding the seal imprint area in the issuer area. Therefore, the issuer element that partially overlaps the imprint area is extracted in a partially missing state.

Further, the journal entry element extraction unit 22 in this embodiment can also extract the issuer element from the FAX information area. The issuer element included in the FAX information area is mainly the FAX number of the issuer (FAX sender).

The business partner database 23 (hereinafter, the database is referred to as "DB") is a so-called business partner master, and the company name, personal name (store name), address, telephone number, fax number, and mail corresponding to the issuer element. Business partner information such as address, department name, person in charge, etc. is stored.

The issuer estimation unit 24 collates the issuer element extracted by the journal element extraction unit 22 with the supplier information stored in the supplier DB 23, estimates the issuer of the voucher, and outputs the estimation result. It has a function. Note that the issuer estimation unit 24 may output only the estimation result, or may output the estimation result together with other journal elements extracted by the journal element extraction unit 22.

Specifically, the issuer estimation unit 24 estimates the issuer of the voucher according to the degree of agreement between the extracted issuer element and the supplier information of the supplier DB23. That is, even if the issuer element is partially missing, the supplier information having a higher degree of agreement than the other supplier information is estimated as the issuer of the voucher. The estimation result is not necessarily limited to one issuer, and a plurality of issuer candidates may be listed in accordance with the rank having a high degree of coincidence. Further, the estimation result may be accompanied by information on the degree of matching indicated by, for example, a percentage display, in addition to the supplier information.

Further, the issuer estimation unit 24 can collate with the supplier information including the issuer element extracted from the FAX information area. Since the issuer information of the FAX information area is located at the end of the voucher, it is unlikely to be lost due to imprint or the like, and a higher degree of matching can be obtained.

The display unit 12 is, for example, a display of the accounting processing device 10, and has a function of displaying the journal elements extracted by the journal element extraction unit 22 and the estimation result of the issuer estimated by the issuer estimation unit 24. There is. The journal element and the estimation result of the issuer are transmitted not only to the display unit 12 of the accounting processing device 10 but also to the user's information terminal 4, so that the user can also check the extraction result of the journal element and the estimation result of the issuer. You can

Next, the journal element determination unit 13 has a function of performing a confirmation process for determining the journal element displayed on the display unit 12 by the service provider or the program for determining the journal element. That is, in the journal element determination unit 13, for example, the person in charge on the service provider side confirms the journal element displayed on the display unit 12 and the estimation result of the issuer, and confirms the journal element without any problem as it is. , For the problematic journal element, correct it and then perform the confirmation operation. Further, when a plurality of issuer candidates are listed as the estimation result of the issuer estimation unit 24, the person in charge selects and determines one issuer candidate in the journal element determination unit 13.

The journal unit 14 has a function of outputting an account item corresponding to the journal element determined by the journal element determination unit 13. The output of the account item corresponding to this journal element is performed by, for example, the journal AI. The journal AI is an automatic journal AI that has learned in advance to output an account item for a journal element by machine learning in the learning system 15. 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 area identification AI, the journal AI, and the text AI, and supplying the learned AI. Specifically, the learning system 15 generates journal entry element extraction AI by performing machine learning (so-called deep learning) based on teacher data (learning data) including the voucher image data and journal entry elements included in the image data. .. Further, the learning system 15 generates a journal AI by performing machine learning based on teacher data (learning data) composed of a journal element and an account item corresponding to the journal element.

In the accounting processing device 10 configured as described above, the voucher analysis unit 11 extracts journal elements from the voucher image data and converts them into texts, and particularly identifies the imprint area for the voucher including the imprint. , The issuer area that overlaps with the imprint area is identified, the issuer element is extracted from the issuer area, and the issuer of the voucher is estimated by collating with the customer information of the customer DB23.

Here, referring to FIG. 2, a flowchart showing the flow of voucher analysis executed by the voucher analysis unit 11 is shown, and the voucher analysis method will be described in detail below with reference to the same flowchart. The voucher analysis starts when the image acquisition unit 20 receives the voucher image data from the user.

First, in step S1, the voucher analysis unit 11 identifies a region corresponding to a journal element from the voucher image data by the region identification unit 21.

Then, in step S2, the journal element extraction unit 22 extracts the journal elements in the identified area and converts them into text.

Further, in step S3, the voucher analysis unit 11 determines whether or not the imprint region is identified in the image data of the voucher by the region identification unit 21. If the determination result is false (No), that is, if the received image data is a voucher without imprint, the process proceeds to step S4.

In step S4, the voucher analysis unit 11 outputs the text of the journal element extracted by the journal element extraction unit 22 to the display unit 12 and ends the routine.

On the other hand, if the determination result in step S3 is true (Yes), that is, if the voucher has an imprint such as an invoice or a receipt, the process proceeds to step S5.

In step S5, the voucher analysis unit 11 identifies the imprint area in the image data of the voucher by the area identification unit 21, and identifies the issuer area that overlaps with the imprint area (area identification step).

Then, in step S6, the journal element extraction unit 22 extracts the issuer element from the issuer area (issuer element extraction step).

Further, in step S7, the voucher analysis unit 11 determines whether or not the fax information is described in the FAX information area in the image data of the voucher identified by the area identification unit 21. If the determination result is false (No), the process proceeds to step S9, and if the determination result is true (Yes), the process proceeds to step S8.

In step S8, the journal element extraction unit 22 extracts the issuing source element (for example, FAX number) in the FAX information area, and the process proceeds to step S9.

In step S9, the voucher analysis unit 11 collates the issuer element extracted in steps S6 and S8 with the supplier information stored in the supplier DB 23 by the issuer estimation unit 24 (issuer). Estimation process).

In step S10, the voucher analysis unit 11 outputs the supplier information having a high degree of agreement with the extracted issuer element to the display unit 12 as the issuer estimation result by the issuer estimation unit 24 (issuer estimation step). .. In step S10, the voucher analysis unit 11 also outputs the text of the journal element other than the publisher element extracted by the journal element extraction unit 22 to the display unit 12 as in step S4, and the routine To finish.

In this way, the voucher analysis unit 11 extracts and displays the journal element including the publisher element from each image data by executing the routine every time the image acquisition unit 20 acquires the image data of the voucher. It can be output to the unit 12.

Next, referring to FIG. 3, an explanatory diagram for estimating the issuer from the invoice including the imprint is shown, and a specific procedure for estimating the issuer from the invoice will be described based on the diagram.

When the image data P of the invoice as shown in the left part of FIG. 3 is received by the image acquisition unit 20, the area identification unit 21 identifies the area corresponding to the journal element as shown by the dotted line. For example, an area containing a "invoice" indicating the type of voucher, a "subject" corresponding to a description, a "billing date" corresponding to a date, and a "total amount" corresponding to an amount is identified.

Further, the voucher analysis unit 11 identifies the imprint area S by the area identification unit 21 and identifies the issuer area A overlapping with the imprint area S. Further, the journal element extraction unit 22 extracts the publisher element in the publisher area A as text. In the case of FIG. 3, the issuer elements extracted here are “ABCDEFG Co., Ltd.”, “〒100-100”, “Kasumigaseki, Chiyoda-ku, Tokyo”, “TEL:03-1234”, and “FAX:03”. -9876-5432".

Further, the voucher analysis unit 11 identifies the FAX information area F by the area identification unit 21, and the journal element extraction unit 22 prints the FAX information area F in the FAX information area F, ie, “FAX number: 03-9876-”. 5432” is extracted as text.

Then, the issuer estimation unit 24 collates these extracted issuer elements with the supplier information stored in the supplier DB 23, and outputs the supplier with a high degree of matching as the issuer estimation result. Since the issuer estimation result is based on the supplier information stored in the supplier DB 23, it is output as information without any loss, and the information is utilized for journal entry by, for example, the journal unit 14.

In the case of FIG. 3, "ABCDEFG Co., Ltd.", "Kasumigaseki, Chiyoda-ku, Tokyo" and "TEL: 03-1234-" extracted from the publisher area A were partially missing from the supplier information. Although it is information, the degree of matching is higher than that of other business partner information. Further, even in the issuer area A, since the FAX number can be extracted in the FAX information area F without any loss, the degree of coincidence becomes particularly high depending on the FAX number.

Since "ABCABC Co., Ltd." in the supplier information also partially matches the supplier name, address, telephone number, and FAX number, for example, the information on the "ABCABC" and "ABCDEFGH" companies is also included. Alternatively, a plurality of candidates may be output as the issuer estimation result. Further, in this case, the degree of agreement of each candidate may be quantitatively indicated by a percentage display or the like.

As described above, in the accounting processing system 1 including the voucher analysis unit 11 according to the present embodiment, the journal element area including the journal element is identified from the image data of the voucher, and the issuer indicating the issuer of the voucher is issued from the journal element area. The issuer can be estimated by extracting the original element and collating it with the supplier information of the supplier DB23.

In addition, the issuer is estimated according to the degree of matching between the issuer element and the supplier information, so that, for example, imprints, foreign substances, handwritten characters, etc. are covered with the issuer information, or the voucher itself is damaged. , Even if the characters of the issuer information are faint and the issuer information is partially missing, it is possible to estimate and output the accurate issuer element that compensates for the loss from the supplier information.

In particular, in the present embodiment, by identifying the imprint area and identifying the issuer area that overlaps the imprint area, the issuer element is included even when a plurality of company names are described. It is possible to quickly recognize the area to be covered.

Further, the voucher analysis unit 11 identifies the fax information area in the image data of the voucher by the area identification unit 21, and extracts the issuer information from the fax information area to more reliably estimate the issuer. it can.

In this way, by being able to recognize the issuer of the voucher more quickly and accurately from the image data of the voucher, it is possible to reduce the manual confirmation and input of the issuer of the voucher, and it is possible to reduce the input work of the issuer. The efficiency can be improved.

Then, by performing the journal entry by the journal entry unit 14 based on the journal entry element of the voucher including the issuer element accurately estimated in this way, the accounting processing system 1 can realize an accurate journal entry.

The above is the description of the embodiment of the present invention, but the aspect of the present invention is not limited to this embodiment.

For example, in the above embodiment, the invoice shown in FIG. 3 has been described as an example, but the voucher is not limited to the invoice, and can be similarly applied to a voucher including other imprints such as a receipt. ..

Further, in the above embodiment, the journalizing unit 14 uses the journalizing AI to perform automatic journalizing, but the journalizing may be performed by an automatic journalizing program that does not use AI.

In the above embodiment, the issuer information extracted from the issuer area is included in the issuer element extracted from the FAX information area to be compared with the supplier information, but the issuer information is not limited to this. Other information stored in the business partner DB may be used instead of the information stored in the business partner DB.

For example, the voucher analysis unit may identify the account information area B including the bank account information of the transfer destination shown in FIG. 3, and extract the bank account information of the account information area as the issuer information.

Further, the voucher analysis unit may identify the number information area including the billing number (invoice No.) shown in FIG. 3, and extract the billing number of the number information area as the issuer information. Such number information is not limited to the billing number, and may be an estimate number if it is a quotation and an order number if it is a purchase order. In addition to this, numbers such as a corporate number, a business establishment number, and a transaction code may be used as issuer information.

Further, not limited to the example of FIG. 3, the voucher analysis unit may identify the mail address information area including the mail address information and extract the mail address information of the mail address information area as the issuer information.

In this way, by using the issuer elements in areas other than issuer area B, the issuer can be more accurately estimated.

1 Accounting Processing System 2 Communication Network 3 Reading Device 4 Information Terminal 10 Accounting Processing Device 11 Voucher Analysis Unit (Voucher Analysis Device)
12 display section 13 journal element determining section 14 journal section 15 learning system 20 image acquiring section 21 area identifying section 22 journal element extracting section (issuer extracting section)
23 Business partner database (storage unit)
24 Publisher estimation section

Claims

An area identification unit that identifies the journal element area including the journal element from the image data of the voucher,
An issuer element extraction unit that extracts an issuer element indicating the issuer of the voucher from the journal element region identified by the region identification unit;
A storage unit that stores customer information,
An issuer estimator that estimates the issuer of the voucher by comparing the issuer element extracted by the issuer element extractor with the supplier information stored in the storage;
Voucher analyzer equipped with.
Claim 1 that the issuer estimation unit estimates the issuer of the voucher according to the degree of agreement between the issuer element extracted by the issuer element extraction unit and the business partner information stored in the storage unit. The described voucher analyzer.
The area identification unit identifies an area of a seal imprint from the image data of the voucher, and identifies an issuer area including an issuer element that overlaps the area of the imprint,
The voucher analysis apparatus according to claim 1, wherein the issuer element extraction unit extracts the issuer element from the issuer area identified by the area identification unit.
The area identification unit can identify the FAX information area in which the transmission information is described at the time of receiving the FAX as the journal element area.
The issuer element extraction unit can extract the issuer element from the FAX information area identified by the area identification unit.
The voucher according to any one of claims 1 to 3, wherein the issuer estimation unit compares the issuer element extracted from the FAX information area with the supplier information stored in the storage unit. Analyzer.
The area identification unit can identify an account information area in which bank account information is described as the journal element area.
The issuer element extraction unit can extract the issuer element from the account information area identified by the area identification unit.
The voucher according to any one of claims 1 to 4, wherein the issuer estimation unit compares the issuer element extracted from the account information area with the supplier information stored in the storage unit. Analyzer.
The area identification unit can identify a number information area in which unique number information unique to a business partner is described as the journal element area.
The issuer element extraction unit can extract the issuer element from the unique number information area identified by the area identification unit.
6. The issuer estimation unit collates with the supplier information stored in the storage unit, including the issuer element extracted from the unique number information area, according to claim 1. Voucher analyzer.
The area identification unit can identify the e-mail address information area in which the e-mail address information is described as the journal element area.
The issuer element extraction unit can extract the issuer element from the e-mail address information area identified by the area identification unit.
7. The issuer estimation unit compares the issuer element extracted from the email address information area with the supplier information stored in the storage unit, according to claim 1. Voucher analyzer.
The voucher analysis device according to any one of claims 1 to 7.
And a journalizing section that outputs an account item corresponding to the journalizing element of the voucher including the issuer element.
By computer
An area identification process that identifies the journal element area including the journal element from the image data of the voucher,
An issuer element extracting step of extracting an issuer element indicating the issuer of the voucher from the journal element area identified by the area identifying step;
An issuer element extraction step of estimating the issuer element of the voucher by collating the issuer element extracted by the issuer element extraction step with the supplier information stored in the storage unit;
A voucher analysis method for executing.
A voucher analysis program for causing a computer to execute the voucher analysis method according to claim 9.