JP2017054469A - Data management system, method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make a document, such as a contract, voucher, or receipt, into an image, and thereby to simplify confirmation of material for an internal audit, and a confirmation task of material for an accounting audit by checking journalization with an original voucher.SOLUTION: A data management system searches for a picture image on the basis of a picture image, such as a document like a contract, voucher, or receipt, and a keyword of transaction contents, and presents a user with the picture image together with the keyword in a ledger report format as a search result, and thereby the user can intuitively understand the contents of the picture image and search for a related picture image.SELECTED DRAWING: Figure 3-1

Description

The present invention is the origin of journal entries (transactions) in accounting processing when using a filing system that visualizes vouchers used together with accounting processing in accounting firms, accounting firms' consulting companies, general corporations, etc. The present invention relates to a technique for searching for a proof image of related documents (contracts, receipts, receipts, etc.) based on the documents (contracts, receipts, receipts, etc.).

Conventionally, documents related to tax accounting are read and imaged with a scanner, text data obtained by performing OCR processing is associated with an image as a search keyword, and the image of the read document is sorted and stored in a specific folder, There is a technique for searching document images by text search. (For example, Patent Document 1)

  In addition, there is a technique for associating journal data entered in a journal with an image such as a receipt that is the subject of the journal and using it for auditing. (For example, Patent Document 2)

JP 2002-269126 A JP200413353569 Patent 5243054

In the technique disclosed in Patent Document 1, since text data obtained by reading a document or the like with a scanner and performing OCR processing is used as a search keyword, the keyword is not generated according to a specific rule necessary for accounting work. . Therefore, while it is suitable for ambiguous text search, there is a problem that it is not suitable for search using conditions specific to accounting (amount, date, account item, etc.).

Further, in the technique of the above-mentioned Patent Document 2, it is possible to search and confirm a receipt image associated with the document data. However, since the document data is based (premise), there is no document data, There is a problem that it is not possible to search for a further receipt image related to the receipt image or a related contract image.

In the technique disclosed in Patent Document 3, a document or the like can be read and converted into an image, and an OCR process can be performed to sort the read image. There is a problem that is not suitable for. Also, there is a file management system for managing the contract contents, image and contract amount in the prior art.

The present invention has been made in view of the above, and an image of a document (contract, receipt, receipt, etc.) that originated an occurrence of a journal (transaction) in accounting processing, transaction details, etc. are specific to accounting processing. The texts created in accordance with the rules (the following are accounting rules, which are rules based on accounting processes such as subject, date, amount, etc.) are integrated and managed, and related documents (contracts, receipts, receipts, etc.) images and Search for journals (transactions), display documents (contracts, receipts, receipts, etc.) in the form of books in the form of images, and visually check them to assist and input journals intuitively. The purpose is to support later verification.

In order to solve the above-described problem, the data management system according to the present invention which is the first invention accepts input of text information for accounting processing regarding a testimony image, adds it to the image as image modification information, A data management system that searches for testimony images using image modification information and enumerates the testimony images according to the format (in order of items such as books and financial statements) set based on accounting rules,
Accepting input (or acquisition) for each element (Note 1) set based on accounting rules for the displayed testimonial image, and modifying the testimonial image with this input (or acquisition) text information Image modification information recording means for adding and recording as image modification information;
Search means for searching for a testimony image using image modification information;
Specify whether to enumerate on the basis of a specified book format including at least one of the ledger format, trial balance format, and journal format as the format for enumerating the testimony images obtained by searching Image enumeration style designation means for
The evidence images obtained by searching are listed according to the selected book format (in the order of items in the book).

In addition, the data management system according to the present invention as the second invention provides, for each element, as text information when inputting (or acquiring) each element set based on the accounting rule with respect to the above-described test image. Is equipped with a question type input acceptance means that presents a set of questions and answer options,
A query type input-based image modification information recording means for recording a set of a question and a selected answer option as image modification information for modifying the testimony image is provided.

In addition, the data management system according to the present invention as the third invention provides a book or slip as text information when inputting (or acquiring) each element set based on the accounting rule with respect to the above-described test image. For each element, there is an accounting data input receiving means for receiving data input,
An accounting data input-based image modification information recording means that records input data for each received element as image modification information for modifying the display image,
It is characterized by having.

In addition, the data management system according to the present invention as the fourth invention relates to the above-mentioned test image as text information when inputting (or acquiring) each element set based on the accounting rule. It has text information recognition acquisition means to acquire meaningful text information on accounting rules by performing character recognition pre-processing on the image, and the acquired text information is used as image modification information for modifying the test image. Image modification information recording means based on character recognition preprocessing to be added and recorded is provided.

In addition, a system for enumerating the image of a test according to the present invention as a fifth aspect of the present invention is a system for searching and enumerating another image of a test image related to a predetermined image of the test image. When a specific ID image is selected from among them, the image modification given to the specific ID image is used as a means of searching for and listing other ID images related to the selected ID image. Based on accounting rules (including the same month in the previous year and amount range) for each element of interest and the element of interest designating that specifies which element of information (summary character string, date, amount of money, etc.) ) Search range condition or search range condition that is set as a search range condition (including switching from the subject to the other subject) and stored in the storage unit as accounting-specific processing A storage means; a means for searching for image modification information recording means on the basis of the designated element of interest; and a search range condition corresponding to the element of interest; and other means related to the specified evidence image It is characterized by searching and enumerating testimony images.

In addition, in the certificate image enumeration system according to the present invention which is the sixth invention, as the element-of-interest specifying means, means for automatically acquiring the element of interest by indicating a predetermined item or neighborhood of the certificate image, or separately Means for displaying and selecting a list of element items, and for this element-of-interest specifying means, and for each element of interest, the search range condition (including the same month in the previous year and amount range) or accounting Search range condition storage means for setting a search range condition and storing it in the storage unit (including switching the subject to be searched from the subject to the other subject) as a specific process, and related to the specified test image It is characterized by searching and enumerating other evidence images.

A certificate image enumeration system according to the present invention as a seventh invention is characterized in that journal data is created by combining image modification information of certificate images obtained by searching.

In addition, the system for enumerating testimony images according to the present invention as the eighth invention is based on the testimonial image obtained by searching while switching each book or the appropriateness of the accounting process in each book based on the image modification information. It is characterized by having a means for confirming.

Further, according to the ninth aspect of the present invention, there is provided a method for enumerating a testimony image, accepting input of text information for accounting processing regarding the testimony image, adding it to the image as image modification information, and adding the image modification information to the image. This is a method for enumerating the testimony images according to the format (order of items such as books, financial statements, etc.) set based on accounting rules by searching for the testimony images, and storing the testimony images An input (or acquisition) is received for each element set based on the accounting rule with respect to the displayed testimony image from the DB obtained and presented from the database. Image modification information recording step for adding and recording the text information as image modification information for modifying the display image The search step for searching for a testimony image using image modification information and the format for enumerating the testimony images obtained by the search are at least one of the ledger format, trial balance format, and journal format. An image enumeration format designating step for designating whether to enumerate based on a predetermined book format including one or more book formats, and the image obtained by searching is selected as the selected book format (book (In order of items).

In addition, the method for enumerating testimony images according to the present invention as a tenth invention is a method for searching and enumerating other testimony images related to a predetermined testimony image, wherein a plurality of testimony images are listed. When a specific ID image is selected from among them, image modification given to the specific ID image is performed as a process of searching for and listing other ID images related to the selected ID image. Specifying which element of information (summary character string, date, amount of money, etc.) is to be searched for, the element-of-interest specification step, and for each element of interest, based on accounting rules (including the same month in the previous year and amount range) ) Search range condition or search range condition that is stored in the storage unit by setting the search range condition as a process specific to accounting (including switching the subject to be searched from the subject) And a step of searching for the image modification information recording means based on the designated attention element and the search range condition corresponding to the attention element, and other steps related to the specified evidence image It is characterized by searching and enumerating testimony images.

An eleventh image enumeration system according to the present invention, which is an eleventh aspect of the present invention, is a system in which accounting data input with respect to a testimony image and the testimonial image are associated and recorded in a storage area. Extract each element of the associated accounting data, add it to the image as image modification information, search for the test image using the image modification information, and set the format (book, financial) This is a system that enumerates testimonial images according to the order of items such as tables, etc., and extracts the elements that make up each accounting data and puts each element into a file of the testimonial image associated with the accounting data. On the other hand, the image modification information derived from the accounting data given as the image modification information for modifying the test image And at least a ledger format, trial balance table format, journal book format as a format for enumerating the testimony images obtained by searching Image enumeration format designating means for designating whether to enumerate based on a predetermined book format including one or more book formats, and the selected image format is obtained as a result of the search. They are listed according to the order of items in the book.

In addition, an image enumeration system according to the present invention as a twelfth aspect of the present invention is a system in which accounting data input with respect to an identification image and the identification image are associated with each other and recorded in a storage area. Search for other related images using each element, and enumerate other related images according to the format (in the order of items such as books and financial statements) set based on accounting rules. And a search means for searching for other testimony images related to a predetermined testimony image using each element constituting the accounting data, and other testimony images obtained by the search. As a form when enumerating, based on a prescribed book format including at least one of the ledger format, trial balance table format, journal format Image enumeration format designating means for designating whether to enumerate, and enumerate the other related evidence images obtained by searching according to the selected book format (in the order of the items in the book) It is characterized by.

A system for enumerating testimony images according to a thirteenth aspect of the present invention is a system for searching and enumerating other testimony images related to a predetermined testimony image, wherein a plurality of testimony images are listed. When a specific ID image is selected from among them, the image modification given to the specific ID image is used as a means of searching for and listing other ID images related to the selected ID image. Focusing element designating means for specifying which element (summary character string, date, amount, etc.) of the accounting data associated with the information or certificate image is to be searched and accounting rules for each focused element As a search range condition or accounting-specific processing (including switching from the subject to the other subject) Based on the search range condition storage means for setting the search range condition and storing it in the storage unit, the designated element of interest, and the search range condition corresponding to this element of interest, image modification information recording means or accounting data is stored. Means for searching for each constituent element, and searching for and listing other testimony images related to the specified testimonial image.

In addition, in the hail image enumerating system according to the fourteenth aspect of the present invention, as the element of interest designating means, means for automatically acquiring the element of interest by indicating a predetermined item or neighborhood of the image of hail image, or separately Means for displaying and selecting a list of element items, and for this element-of-interest specifying means, and for each element of interest, the search range condition (including the same month in the previous year and amount range) or accounting Search range condition storage means for setting a search range condition and storing it in the storage unit (including switching the subject to be searched from the subject to the other subject) as a specific process, and related to the specified test image It is characterized by searching and enumerating other evidence images.

In addition, the certificate image enumeration system according to the present invention, which is the fifteenth invention, combines image modification information of the certificate image obtained by searching or elements of accounting data associated with the certificate image. The journal data is created.

The certificate image enumeration system according to the present invention, which is the sixteenth aspect of the present invention, is an accounting image associated with the certificate image or its image modification information or certificate image obtained by searching while switching each book. A means for confirming the appropriateness of the accounting process in each book based on the data is provided.

In the present invention, questions and answers for obtaining keywords specific to accounting and image images such as documents (contracts, receipts, receipts, etc.) that originated are presented to the user in a list format. Can intuitively verify the testimonials using image images, and can also be used as a reference when entering transactions, so even those who have accounting knowledge such as bookkeeping but lack knowledge of handling image filing systems can be It can be searched, and verification and input of test data can be performed easily. In addition to managing contract-related items such as the contract content and image and contract amount as in the prior art, the contract content as a comment on the image, transaction content generated by transactions after the contract other than the contract amount, etc. Can be added.

FIG. 1 is a diagram showing a hardware configuration of a consultant terminal in a data management system according to the present invention. FIGS. 2-1 to 2-4 are diagrams showing some configuration examples of the data management system according to the present invention. FIG. 3-1 is a diagram showing a part from image reading to keyword generation registration of the entire flow in the data management system according to the present invention. FIG. 3-2 is a diagram showing an image search portion of the overall flow in the data management system according to the present invention. FIG. 4A is a diagram illustrating a detailed flow of query formula input (documents such as contracts). FIG. 4-2 is a diagram showing a detailed flow of query type input (certificates such as a receipt). FIG. 4-3 is a diagram illustrating a detailed flow of journal entry while viewing an image such as a receipt. FIGS. 5-1 to 5-15 are diagrams showing table configurations in the data management system according to the present invention. FIG. 6A is a diagram illustrating a display screen for query type input (documents such as contracts). FIG. 6B is a diagram illustrating a display screen for inputting a question formula (a voucher such as a receipt). FIG. 6-3 is a diagram illustrating a display screen for journal entry while viewing an image such as a receipt. FIG. 6-4 is a diagram showing a display screen for setting search conditions. FIG. 7-1A is a diagram in which a display screen of a keyword search result of an image input with a query formula is displayed in a ledger format. FIG. 7-1B is a diagram showing a pop-up screen on which an image is selected and a question and answer of a question type input are displayed. FIG. 7-1C is a diagram showing a display screen of a keyword search result of an image input with a query formula in a trial calculation table format. FIG. 7-1D is a diagram illustrating a display screen for the result of the related search. FIG. 7-2A is a diagram showing a display screen of keyword search results of an image input as a journal. FIG. 7B is a diagram illustrating a search condition setting pop-up display screen when an image is selected and a related search is performed. FIG. 8A is a diagram illustrating a flow of processing for searching for related images. FIG. 8B is a diagram illustrating a flow of processing for searching for related images. FIG. 9A is a diagram illustrating the structure of the file storage unit. FIG. 9-2 to FIG. 9-6 and FIG. 9-8 to FIG. 9-10 are diagrams showing the contents of tables and dictionaries registered in the table storage area. 9-7 and 9-11 are diagrams showing the contents of the search index registration area. FIG. 10 is a flowchart of the document image registration process when the registration destination is selected in advance by the operator. FIG. 11 is a flowchart of the image data registration process when the registration destination is not selected in advance. FIG. 12 is a diagram showing an example of a screen when an error ends in the flowchart of FIG.

In the data management system of the present invention, there are roughly two embodiments. In the first embodiment, using various keyword assignment methods, keywords according to accounting-specific rules are assigned to images of documents (contracts, receipts, receipts, etc.), and related images are searched and referenced. Is the case. As a second embodiment, an image to which a keyword is assigned is displayed in the form of a book table, a related image is searched, intuitive input assistance mainly based on the image from the search result, journal verification, etc. This is a case where it is carried out and will be described separately.

<Overview of overall processing flow>
First, with reference to FIGS. 3A and 3B, an overview of the flow of information and processing in the data management system of the present invention will be described. As used herein, “certification” refers to the collective name of the documents that originated the accounting process. Specifically, in addition to receipts and receipts, it is not necessarily a one-to-one relationship with journals. Lease contracts, sales contracts, license contracts, maintenance contracts, etc.), insurance policies, delivery notes, purchase slips, deposit slips, order forms, and the like.

In the present invention, a certified public accountant office, a tax accountant office (hereinafter, simply referred to as “accounting office”) and its advisors as a pre-processing, or receipt of a certificate of receipt, contract, etc. from the advisor. The above-mentioned receipts, receipts, contracts, etc. are imaged with a mobile device such as a scanner or a smartphone (hereinafter abbreviated as a scanner) (hereinafter referred to as a “proof image”). A testimonial image is created and registered in the image management table (image DB) (FIGS. 10 and 11). The document image registration process will be described later.
Then, a test image that is an image of a document is read (S3-1-1 to S3-1-3), and input of text information for accounting processing (a plurality of input methods) is accepted (S3-1-4). ~ S3-1-7), the inputted text information is added to the test image as image modification information and registered in the image management table (image DB) (S3-1-8 to S3-1-10).
Furthermore, the image modification information is used to search for the test image registered in the image management table (image DB) (S3-2-7 to S3-2-8), and the accounting rule (question and answer set) (S3-2-1 to S3-2-6) in accordance with the format (in the order of items such as books and financial statements) set based on the description as journal elements at the time of manual entry, information on accounting such as dates) ), A system for enumerating proof images (S3-2-9 to S3-2-11).
<About saving documents>
In step S3-1-2, an image of the document is read and stored in the storage unit. At this time, the document is simply stored in a folder for each advisory company, and the accounting office or the like. It can also be sorted and stored in a folder suitably configured for processing by the accounting department of the company.
Hereinafter, the process of allocating to a folder performed in step S3-1-2 will be described.
Note that this folder distribution processing is described as being performed at step S3-1-2, but may be performed when a keyword is assigned to a document image.

<Dictionary handling when storing proofs in a folder with a free keyword dictionary>
FIG. 9A is a diagram illustrating the structure of the file storage unit in the present embodiment. FIGS. 9-2 to 9-6 are diagrams showing the contents of the table and the dictionary 2 registered in the table storage area used in the test image registration process in this embodiment. 9-5 to 9-7 are diagrams showing the contents of tables and dictionaries 1 and 2 registered in the table storage area used in the free keyword search in this embodiment. For free keyword search, INDEX1 (FIG. 9-7), which is an index for free keyword search, is generated based on dictionary 1 and dictionary 2 from table 2 (FIG. 9-3) and table 4 (FIG. 9-8). And register it in the text index registration area. Other processing related to the search is the same as the normal search processing that can be performed by those skilled in the art, and a description thereof will be omitted. 9-8 to 9-11 are diagrams showing the contents of the tables and the accounting dictionaries 1 and 2 registered in the table storage area used in the accounting keyword search in this embodiment. These tables and the dictionary 2 are used by the text search unit and the registration destination determination unit. FIG. 9-2 is a table 1 for associating the upper folder group 201 and the company name shown in FIG. 3 with the management number uniquely set for each customer, the upper folder name, and the company name being associated with each other. . On the other hand, the table 2 shown in FIG. 9C is a table that associates abbreviations or aliases of companies that may be used in individual document data with upper folder names of the table 1, and the management set in the table 1 Numbers are associated with abbreviations or alias lists. For example, in the management number 1 set to “ABC Kogyo Co., Ltd.” managed by the upper folder name “ABC Kogyo” in Table 1, “ABC Kogyo Co., Ltd.”, “ABC Co., Ltd.”, “ABC Tokyo” Possible abbreviations or aliases, such as Similarly, “DEF Trading Co., Ltd.” managed by the management number 2 is associated with an abbreviation or alias such as “D-E F” or “DEF Trading Co., Ltd.”. Even if the official company name (the company name described in Table 1) is not used in the document file, the table 2 associates the document with the abbreviation or alias listed in Table 2 as a document related to the company. It becomes possible.

Next, a table 3 shown in FIG. 9-4 is a table for managing a folder that can be used as a document file registration destination in the file storage unit 23. For each customer management number, there are three types of accounting, tax, and general. It is associated which folder of the lower layer folder 202 can be used. Here, “1” is registered when the folder is available, that is, when the business is undertaken by the customer, and “0” is registered when the folder is not used. In this embodiment, as shown in FIG. 9A, the general folder “DEF Trading” and the accounting folder “GHI Sangyo” are not used. “0” is registered in the “General” column of “equivalent” and the “Accounting” column of management number 3 (corresponding to the GHI industry), and “1” is registered in the other columns.

Information in these tables 1 to 3 is created as a new record when a new customer is registered by a financial processing program operating on a computer system. That is, when an operator registers a new customer in the financial processing program, one upper folder (company name folder) having three lower folders (accounting / tax / general folder) in the folder structure shown in FIG. The management number is set in the table 1, and the company name and folder name of the new customer are registered using the company name of the customer entered by the operator. Also, the abbreviation or alias name of the company is registered in the table 2 by the operator's designation or automatically from the input company name. Here, in order to automatically register abbreviations or aliases, the operator extracts the reading from the input contents before converting to Kanji, or the system has a comparison table of each alphabet letter and katakana notation of the reading in advance. For example, it may be programmed to extract an alias “ABC” from a company name including “ABC”. Further, the information of the table 3 is registered from the contract work information specified by the operator when registering a new customer. These pieces of information are registered in advance in the tables 1 to 3, but the processing is the operation of the financial processing program, and is not directly related to the operation of the data management program of the present invention. Then, since it can be easily understood, further detailed description is omitted in this specification.

FIG. 9-5 and FIG. 9-6 are diagrams showing an example of a dictionary for deriving to which business the document data is related from the keywords included in the document data. The dictionary 1 shown in FIG. 9-5 associates business names (accounting / tax / general) used in the system of this embodiment with document category names related to each business. In the dictionary 2 of FIG. 9-6, a plurality of keywords related to the document category registered in the dictionary 1 are registered. Further, although this is not essential, each keyword is set with a numerical value of weighting according to importance (the numerical value in parentheses in FIG. 9-6).
FIG. 9-9 and FIG. 9-10 are diagrams showing an example of a dictionary for deriving a keyword related to accounting work from the keyword included in the document data. In the accounting dictionary 1 shown in FIG. 9-9, a business name (accounting / tax / general) used in the system of this embodiment is associated with a business category name related to accounting processing among documents related to each business. . Also, in the dictionary 2 of FIGS. 9-10, a plurality of keywords related to the accounting process of the document category registered in the accounting dictionary 1 are registered. Moreover, although this is not essential, each keyword is set with a numerical value of weighting according to importance (the numerical value in parentheses in FIGS. 9-10).
In the accounting dictionary 2, the accounting keywords generated by the comment generation registration processing (S3-2-1) and the accounting processing specific keyword generation registration flag set processing (S3-2-2) may also be registered. Further, the keywords registered in the dictionary 2 may be registered in the accounting dictionary 1 after being classified for each document category.

These dictionaries 1 and 2 and accounting dictionaries 1 and 2 are used when the system or program according to the present invention is shipped, depending on the purpose of use of the computer system to which the data management system, method and program according to the present invention are applied. Alternatively, it is assumed to be registered in advance at the start of system operation. In this embodiment, a computer system for an accounting office is assumed, but it is obvious that the table, dictionary 2, and accounting dictionary 2 used in other industries have other configurations.

<Processing flow for storing the certificate in a folder with a free keyword dictionary>
A processing flow of the data management system configured as described above will be described in detail with reference to FIG. FIG. 10 is a flowchart for explaining an embodiment of the data management method according to the present invention. This figure shows the operation when the registration destination of the target document data is designated by the operator. Here, the document data to be stored is temporarily stored in the document data group storage area of the storage unit in FIG. This document data indicates an electronic file including text data, and may be a file created by a word processor or spreadsheet software, or may be created immediately before by a financial processing program operating in this system.

It may be created in advance by a financial processing program or other software and pooled in a document data group temporary storage area, or may be received via a network from a mail attachment, Internet download, or other terminal in the LAN. . Further, it may be image data acquired from a document reading device (not shown) such as a scanner connected to the system and provided with text information by OCR processing.

In this embodiment, first, the system operator accepts specification of document data and selection of a registration location (step A1). A screen (not shown) presented to the operator from the output unit when selecting document data is displayed in the output unit as a list of document data groups registered in the temporary storage area, and the operator operates the input unit to select one or more items. The file can be selected. This list provides information on file name, format, and creation date. The operator highlights the target file by clicking with the mouse, for example, and presses the confirm button below.

Next, a registration destination selection screen (not shown) is presented to the output unit. Similarly, the operator selects a file registration location and presses the confirm button. The selection and processing instruction unit 11 of the control unit 10 receives this input, extracts a corresponding document file from the document data group, and transfers the processing to the text search unit. In this embodiment, the operator designates only the upper folder 201 for each company shown in FIG. 9-1 and does not designate the business folder below it.

On the other hand, the system checks whether an appropriate company name is specified based on the document content, determines which business folder should contain the document content, and automatically registers the file in that folder. Configured to be. However, the operator may specify a path of two or more hierarchies in step A1, and the system may check all of them.

Next, the text search unit of the control unit 10 performs a text search of the document file using the company name and its abbreviation and alias registered in the tables 1 and 2 shown in FIGS. 9-2 and 9-3 as keys ( Step A2). As a result, if any company name or abbreviation or alias is described in the document file, it is extracted. Specifically, the text search unit searches all of the company names in Table 1 and the list of abbreviations or aliases in Table 2, extracts the management numbers of company names, abbreviations or aliases that have been hit, and stores the search results as registration destinations. Output to the decision unit. The registration destination determination unit determines whether the company selected in step A1 is included in the search result of the document file (step A3).

Specifically, it is compared whether or not the management number such as the hit company name matches the management number of the company selected by the operator. When a plurality of company names are detected in one document file, in this embodiment, if any one of the detected company names matches the selected company name, the determination in step A3 is YES. However, in another embodiment, for example, only the company names that appear most frequently in the document file may be compared. If the company name selected in step A3 is included, the process proceeds to step A5, and if the company name specified by the operator is not included in the document file, the process proceeds to step A4.

If there is no specified company name in the document file, it is determined whether a company name other than the selected registration destination is included (step A4). If there is, the process proceeds to the flowchart of FIG. If not, the process proceeds to step A5.

In step A5, the text search unit again performs a text search for the keyword in the dictionary 2, and the registration destination determination unit determines the document category name based on the result (step A5). Here, there are several methods for determining the document category name. For example, a method of adopting the document category name corresponding to the first hit keyword, a method of adopting the document category name having the largest number of hits, FIG. As shown in the dictionary 2 shown in FIG. 6, weighting information is added for each keyword, and a value obtained by multiplying the number of hits of each keyword by the weighting value is added for each document category name, and the document category name having the largest value is added. The method to adopt etc. can be considered. Furthermore, a search process for accounting processing (S3-2-3 to S3-2-11) may be performed in parallel.

Next, the registration destination determination unit refers to the dictionary 1 in FIG. 9-5 and extracts a business name corresponding to the determined document category name (step A6). The business folder corresponding to the business name extracted here is determined as the folder in which the document file is to be registered. Furthermore, a search process for accounting processing (S3-2-3 to S3-2-11) may be performed in parallel.

Furthermore, the registration destination determination unit checks whether the business folder of the corresponding company can be used with reference to the table 3 in FIG. 9-4 (step A7). If available, the document file is stored in the file storage unit 23 in association with the corresponding business folder in the company name folder selected by the operator (step A8).

In this embodiment, the document category name extracted in the dictionary 2 at this time is adopted as a file name or a part thereof, and is registered in the business folder determined in the dictionary 1. On the other hand, if the determined business folder cannot be used, registration cannot be performed at the registration destination selected by the operator, and the process proceeds to a registration destination candidate presentation process (B: to step B8 in FIG. 11). In another embodiment, as shown in FIG. 12, a screen for prompting re-selection of the registration destination may be displayed, and this input may be repeated from step A1 again.

Next, an example of operation when the registration destination of document data is not selected in advance by the operator will be described using the flowchart of FIG. In the method of this embodiment, candidates for registration destinations are extracted from the text contents of document data for which registration destinations are not designated in advance, and are selected by the operator. First, the selection and processing instruction unit presents the document data group registered in the document data group temporary storage area to the operator, and prompts the user to select document data to be registered (step B1). This process is performed by presenting a temporarily registered document data list screen (not shown) to the operator.

When the file designation is received, the text search unit searches the text of the document data for the keyword in the dictionary 2 of FIG. 9-6, and the registration destination determination unit determines the document category name from the hit keyword (step B2). This is the same as step A5 in FIG. 10, and the category name to which the first hit keyword belongs can be adopted, or the category name having the largest added value can be adopted by assigning a weight to each keyword. .

In the flowchart of FIG. 10, even when the selected company name is not included in the symmetrical document data and another company name is extracted (step A4: YES in FIG. 10), the process is started from here. When there is no selected company name, the company included in the document data is extracted from the system side and proposed to the operator.

Next, the registration destination determination unit refers to the dictionary 1 in FIG. 9-5 and specifies a business name corresponding to the determined category name (step B3). This determines the category name and business name of the document file to be symmetric. Next, the text search unit performs a text search for the keywords in Tables 1 and 2 in FIGS. 9-2 and 9-3 (Step B4). If any company name (or abbreviation or alias) is hit, the process proceeds to Step B6. (Step B5). If there is no company name in the document file (step B5: NO), an error message is notified from the output unit 12 and the process is terminated (step B12).

Next, the registration destination determination unit refers to the table 3 in FIG. 9-4 and confirms whether or not the extracted business folder of the extracted company can be used (step B6). This is performed by confirming whether the record at the corresponding location is “1”. If it is available (step B6: YES), the registration destination determination unit registers the company name in a registration destination candidate list that is presented to the operator later (step B7). If the business folder cannot be used (step B6: NO), the process proceeds to step B8 without being registered in the list.

Next, it is checked whether there is another company name hit in the search using the company name or abbreviation as a key (step B8). If there is a business folder determined for all companies, it can be used. After confirming, register it in the registration candidate list. Also, if the business folder of the company designated by the operator in the process of FIG. 10 is not available (step A7: NO), other company names in the document are extracted in this step B8, and registration destination candidates. Register to the list.

After confirming all company names included in the document file, if company name data exists in this registration destination candidate list (step B9), the list is presented from the output unit 12 to accept selection by the operator (step B10). ).

An example of this screen is shown in FIG. As shown in the figure, a plurality of company names appearing in the document file are presented and can be selected. When the operator selects (highlights) one of the companies and presses the confirm key, the registration destination determination unit associates the symmetric document file with the business folder of the selected company and stores the document file as a file. Save to the department. That is, the determined document category name is adopted as the file name or a part thereof, and this document file is registered in the determined business folder under the selected company name folder. At the time of document file registration, INDEX1 for free keyword search shown in FIG. If there is no data in the list in step B9, that is, if there is no customer company name that can use the business in the document file, it is not known which company's file can be handled, so an error message is displayed. The process ends (step B12).

<Processing after Step S3-1-4 of Overall Processing Flow>
The process of reading the document image and storing it in the storage unit in step S3-1-2 in the overall processing flow (FIG. 3-1) has been described above.
Next, the document image stored in step S3-1-3 of the overall processing flow (FIG. 3-1) is read and displayed, and information (comments, keywords, etc.) relating to the document image is input while referring to the screen or the like. Processes after the process (step S3-1-4) will be described.

<About the input method>
The present invention is based on documents (contracts, receipts), which are the origins of transactions by advisors who have sufficient knowledge of bookkeeping as well as those who have little knowledge of bookkeeping or who have little experience in accounting. In order to be able to assign keywords in accordance with accounting rules to the image of a letter, a receipt, etc., at least one of the following input methods is provided.
(1) Question type input (documents such as contracts)
(2) Question type input (proofs such as receipts)
(3) Journal entry while viewing receipts and other images

(1), (2) is a query formula input, 5W1H (when, where, who, what, why, why) or 5W2H (when, where) based on description rules (specific rules) specific to accounting or tax processing , Who, what, how, why, and how much), the answer is mainly selected, and transaction information is created from the answers that correspond to the questions. Transaction information that is judgment information for selecting a tax category is created, and a keyword is created based on the transaction information.

The question items are the questions shown in the tables of FIGS. 5-9 and 5-10 and the corresponding detailed questions (for example, the item When corresponds to the detail When) in order, and the user answers the question. This completes the transaction information.

(3) Journal entry input while viewing images such as receipts, etc. Documents such as receipts and contracts are displayed on the screen as scanned images, and the user inputs while viewing the images. A keyword is created based on journal data.

<Explanation with the entire processing flow as main and screen display as sub>
From here, referring to FIG. 3-1, FIG. 3-2, it explains along the flowchart of the information and processing in the accounting processing system of the present invention, about the display of the data table and operation screen used during the processing, etc. Will be advanced with explanation as appropriate. FIGS. 3A and 3B mainly explain search index generation / registration processing specialized in accounting. Note that the accounting process and search index generation / registration process using free keywords are conventional processes, and a description thereof will be omitted.

First, the user starts by starting a predetermined document management application (hereinafter referred to as an application). (S3-1-1) After the application is activated, an image of a document to which no search keyword is assigned is read from the image DB in the storage unit (S3-1-2) (here, a receipt (receipt is already received by a scanner or the like). ) And contracts, etc. are imaged and the images stored in the image DB in the storage unit are read. At this stage, the documents are imaged using a scanner or the like. In addition, when an image is captured, an image registration process described later is performed.

  Since the read image is displayed in the application on the PC screen (S3-1-3), the user selects and instructs a desired input method from the following three comment input or keyword assignment methods. (S3-1-4)

When one of the following three input methods is selected, the application reads out the input screen in units of the instructed input method and starts preparation for assigning keywords.
(1) Question type input (documents such as contracts)
(2) Question type input (proofs such as receipts)
(3) Journal entry while viewing receipts and other images

Here, the keyword assignment methods (1) to (3) will be roughly described.
(1) Question type input (documents such as contracts)
“Question formula input (documents such as contracts)” means that the contract contents are used for accounting purposes by selecting or inputting the answers to the questions that are issued in the question format. This is an input method. FIG. 6A is a diagram showing a display screen for inputting a query formula (documents such as contracts). The screen (670) is divided into left and right images read on the left (671), and the query formula on the right. An input screen (672) is displayed. At the time of S3-1-3, the read image (671) is only displayed on the screen, but depending on the input method instructed to select at S3-1-4,
The input screen is read and displayed. Although the screen for inputting a query formula (672) is on the right side here, it may be displayed on a pop-up screen so as to slightly overlap the read image (671), or on the left side, upper side, and lower side.

(2) “Question type input (proofs such as receipts)
“Question type input (vouchers such as receipts)” is a method of inputting transaction contents by selecting or inputting an answer to a question issued in a question format. FIG. 6B is a diagram illustrating a display screen for inputting a question formula (a voucher such as a receipt). The screen (680) is divided into right and left, and an image (681) read on the left side and a screen (682) for inputting a question formula are displayed on the right side. At the time of S3-1-3, the read image (681) is only displayed on the screen, but the input screen is read and displayed according to the input method selected and instructed in S3-1-4. Is done. Although the screen for inputting the question formula (682) is on the right side here, it may be displayed on the pop-up screen so as to slightly overlap with the read image (681), or on the left side, upper side, and lower side.

(3) Journal entry while viewing an image such as a receipt “Entry of journal while viewing an image such as a receipt” is a method of inputting a journal while viewing a read image. FIG. 6-3 is a diagram illustrating a display screen for journal entry while viewing an image such as a receipt. The screen (690) is divided into right and left, and an image (691) read in and a journal entry screen (692) are displayed on the left side. At the time of S3-1-3, the read image (691) is only displayed on the screen, but the input screen is read and displayed according to the input method selected and instructed in S3-1-4. Is done. Although the journal entry screen (692) is on the right side here, it may be on the lower side, or it may be displayed on the pop-up screen so that the journal entry screen (692) slightly overlaps the read image (691). Alternatively, the left side, the upper side, and the lower side may be used.

Next, OCR processing is performed. (S3-1-5) The OCR processing here is an OCR focusing on the date, period, total amount, individual amount, unit price, tax amount, etc. of the items required for the accounting processing for the read image. Processing is performed. OCR processing is executed in the same way as in Japanese Patent Application Laid-Open No. 2015-118488, and recognition processing is performed on characters included in a receipt (receipt) image. If it is a date, “year”, “month”, “day”, and monetary amount are used. If there is, data is acquired based on keywords such as “total” and “current total”.

Then, it is determined whether data has been acquired by the OCR process (S3-1-6). When the data can be acquired (S3-1-6: Y), the process proceeds to S3-1-7 (recognition information addition display on the display screen), and the acquired data is set in the corresponding item column and displayed. For example, when date and amount data can be acquired by the OCR process, the acquired data is set and displayed in the date and amount column. When data cannot be acquired by the previous OCR process (S3-1-6: N), S3-1-7 is skipped and the process proceeds to S3-1-8 to S3-1-10.

<Comment input or keyword assignment for document image>
From here, the input method and process of each method are demonstrated about the comment input thru | or the keyword provision method of above-mentioned (1)-(3).

(1) Questioning process input (documents such as contracts) (S3-1-8) is a processing question when a document or other document such as a contract is described in the accounting rules or tax processing specific description rules (specific rules) ) On the basis of "contract partner, contract purpose, contract period, etc. 5W1H format, etc. And, for each of the large question systems, the question formula input screen selection as a middle concept For example, regarding a large-scale "contract partner" system, as a question for selecting a middle-level question formula input screen, "contract business owner, contract A question (or option) regarding the contract partner such as “Company ...” is prepared, and an input screen (predetermined book) prepared is displayed. Questions are registered in the form of a table (Fig. 5-10) and used to generate a question expression input screen.

On the question formula input screen (FIGS. 6-1 672), the contents of the question are prepared mainly in the form of choices, and an input field in a text format is prepared as an auxiliary. When the user answers a question, a set of questions and answers (options, text input) is recorded in the question DB of the storage unit as a question formula.
The data recorded in the question DB is registered in the accounting side processing questionnaire answer in the question type answer DB after performing processing necessary for accounting processing together with the search processing of the present invention described later.

An answer to each question is created as a comment corresponding to the image, and is given to the image as text data. In the case of a question type input, when an answer is selected for one question, a comment may be generated from the selected contents each time, or when all answers are given to a plurality of questions, all A comment may be generated from the response content. Also, the text data input in the remarks column is added as a part of the comment.

A comment may be just an enumeration of text of a character string or a value that is an answer to a question, but a comment may be created by associating a question with an answer.
Since S3-1-8 is the content indicating the outline process, the detailed process will be described later with reference to FIG.

(2) Questioning method input (Voucher such as receipt) (S3-1-9) When processing question is selected, the general question is “when, where, with whom, what, for what And so on "and so on. For each of the large-scale question systems, a question for selecting a question formula input screen as a middle concept and a detailed question as a subordinate concept are prepared. For example, regarding the “what” system of the large framework, questions regarding the outline of transactions such as “regular payment, accounts payable / unpaid,... If “normal payment” is further selected, an input screen (predetermined book) on which questions (or choices) such as “food, drink, traffic,...” Are prepared is displayed.

In addition, detailed questions such as “Payees, Purchasers,...” Are prepared for the “Where” system of the outline, and “Outside number, who is outside, who is in the company, etc.” Detailed questions such as “...” are prepared, and detailed questions such as “purpose, installation location,...” Are prepared regarding the system of “for what”. The questions are selected as appropriate for each type of book in which transactions are entered, registered in the form of a table (Fig. 5-9), etc., and used when generating a question formula input screen.

On the question formula input screen (FIG. 6-2 682), the contents of the question are prepared mainly in the form of options, and an input field in a text format is provided as an auxiliary. When the user answers a question, a set of questions and answers (options, text input) is recorded in the question DB of the storage unit as a question formula.
The data recorded in the question DB is registered in the accounting-side processing question answer in the question answering DB after performing processing necessary for accounting processing together with the search processing of the present invention described later.

An answer to each question is created as a comment corresponding to the image, and is given to the image as text data. In the case of a question type input, when an answer is selected for one question, a comment may be generated from the selected contents each time, or when all answers are given to a plurality of questions, all A comment may be generated from the response content. Also, the text data input in the remarks column is added as a part of the comment.

A comment may be just an enumeration of text of a character string or a value that is an answer to a question, but a comment may be created by associating a question with an answer.
Since S3-1-9 is the content indicating the outline process, the detailed process will be described later with reference to FIG.

(3) Processing when journal entry input (S3-1-10) is selected while viewing an image such as a receipt The user inputs a journal entry while viewing the image. The input items include date, amount, and account. Items such as subjects, descriptions, tax zones, and divisions are input from a keyboard or the like to complete a journal entry and are registered in the journal data DB.
Also, the input data entered in the journal entry is created as comment data separately from the journal data, and is given to the image as text data. If journal data has already been created, comment data may be created by extracting from the journal data. FIG. 5-7 shows the basic data structure table format of journal data.
Since S3-1-10 is the content showing the outline process, the detailed process will be described later with reference to FIG. 4-3.

The comment data output in the above (1) to (3) is added as image modification information to the image data, and when the image data is read from the image DB, it is overwritten and saved in the image DB. When the image data is not stored in the image DB (when the image data is read by the scanner), the image data is newly stored. Note that the answers and transaction data input in S3-1-8, S3-1-9, and S3-1-10 are processed in S3-1-8, S3-1-9, and S3-1-10. As shown by the arrow leading to “accounting process” on the right side of FIG. 3-1, the data is transferred to the accounting process side (question type answer DB and journal data DB), journal data is generated, and accounting process is performed. Is done. In addition, the responses and transaction data input in S3-1-8, S3-1-9, and S3-1-10 as processes other than those in FIG. 3-2 are S3-1-8, S3-1-9, An index for free keyword search is created from each processing block of S3-1-10, is registered in the text index registration area, and processing related to free keyword search is also performed.

<Extracting accounting-specific keywords from comment acquisition / generation and flag set processing>
Here, the process starting from the S3-2-1 comment generation registration process will be described with reference to FIG.
In the S3-2-1 comment generation / registration process, comments are generated by aggregating the comment data output at each processing stage for each of the cases (1) to (3) described above to generate a table 4 (FIG. 9-8). ).

<Outline of accounting-specific keyword extraction and flag set>
Next, the process proceeds to the accounting process specific keyword generation / registration process (S3-2-2).
From the comment generated and registered in S3-2-1, referring to the accounting dictionary 2 (FIG. 9-10) in which keywords used in accounting processing are registered, the accounting-specific keyword is used as an accounting-only search index. The accounting keyword is generated and registered in INDEX2 (FIGS. 9-11).

A keyword can be simply a text string or text of values, but by referring to the accounting dictionary 2, if the keyword is an advertising expense, information about whether it is a debit item or a credit item and transaction If you want to confirm the proof image related to advertising expenses by adding information about the other subject, or if you search by subject in the free keyword search, not only the item you want to see (the advertising expense is a debit item), but also an unnecessary proof Whereas hail images (advertising costs are credited items) are searched, accounting keyword search only searches for hedging images whose advertising costs are debited items, and efficiently confirms only the hedging images required for accounting processing. Can do.

In addition, in the case of (1) question type input (documents such as contracts) and (2) question type input (vouchers such as receipts), a keyword combining a question and an answer may be generated. (3) In the case of journal entry while looking at an image such as a receipt, a keyword combined with each item (date, amount, account item, etc.) necessary for transaction input may be generated.

If another accounting dictionary or table cannot be prepared as another embodiment, a keyword is generated as a search index from the comment generated and registered in S3-2-1. The keyword is simply a character string. You can simply enumerate the texts of values and values, but in the case of (1) Question type input (documents such as contracts), (2) Question type input (certificates such as receipts) A combined keyword may be generated. (3) In the case of journal entry while viewing an image such as a receipt, a keyword combined with each item (date, amount, account item, etc.) of journal data may be generated.

The keyword extraction dictionaries are broadly divided into free keyword dictionaries used in conventional document management systems, and accounting registered according to accounting-specific rules sufficient to finalize accounting processing. -Two dictionaries are prepared: a keyword dictionary.
Further, a flag for discriminating an accounting keyword (accounting process unique keyword) and a free keyword may be prepared and managed in one table.

The generated keywords are extracted from the comment data stored in the image DB as image modification information and necessary comments are extracted and registered in the table 4 (FIGS. 9-8), and then the accounting dictionary 2 (FIG. 9). An index for searching for an accounting keyword is generated based on 9-9) and registered in INDEX2 (FIGS. 9-11). Further, when generating the accounting keyword, not only the accounting dictionary 2 but also the accounting dictionary 1 in which the accounting keyword classification is created may be referred to.

In another embodiment, the generated keyword is added as a search index to the image data to which the comment data stored in the image DB is added as image modification information, and the corresponding image data is overwritten and stored. . FIG. 5-8 shows data in which image data, comment data, and a search index are integrated in a table format.

<Process to extract accounting-specific keywords from comments>
The keywords for the above-described search index are set as keywords to be searched preferentially at the time of subsequent search execution, so that the accounting dictionary 2 in FIG. 9-10 in which keywords specific to accounting processing in the dictionary DB are registered is used. With reference to the weight, S3-2-2 “Set flag to keyword unique to accounting process” is performed.

In the accounting dictionary 2, keywords and weighting information are associated with each other as in the following example, and by referring to this association, keywords specific to accounting are extracted from the comments as search indexes. The weighting information is set so that the higher the numerical value, the higher the search priority.

Note that the method of using the accounting dictionary 2 when acquiring accounting-specific keywords is not limited to this, and there is a character string that hits the keyword dictionary several times instead of or together with the weighting information. Whether the number of hits is large or not may be used as a criterion for extraction.
<Example of dictionary DB>
Keyword weighting information
Travel expenses Transportation expenses 0.5
Bus fee 0.4
Meeting 0.2
Consumption tax 0.7
Accounts receivable 0.6

In the above explanation, it was explained that a comment-specific keyword was extracted from the comment once it was obtained (input), but information related to the proof image was obtained (input) by referring to the proof image. ), It may be directly acquired (input) as a keyword.
For example, in the case of a query type input, the question is often directly related to directly acquiring an accounting-specific keyword.

<Overview of related document image (certification image) search processing>
From here, it is optimized for the accounting process using the image data, comment data, and accounting keyword search index (FIGS. 9-11) generated in the processes from S3-1-1 to S3-2-2. The search process will be mainly described.

In this embodiment, the search index is separately provided, but the search processing may be performed on data in which image data, comment data, and search indexes (accounting keyword search, free keyword search) are integrated.

The user starts by starting a search subprogram from a predetermined document management application (hereinafter referred to as an application). (Not shown), the application reads a search selection screen from the program storage unit in the storage unit (S3-2-3), and displays a screen for receiving search conditions and the like in the application screen (S3-2-4). ) Then, for the input of display method and detail display selection (S3-2-5) and image processing image search condition setting for accounting processing (S3-2-6), necessary items are selected and input from the keyboard or the like.

In a conventional document management application, keywords are automatically extracted from text information of a document file (image) by a fuzzy search such as a partial match search with a free keyword and listed as a comment (Table 2 in FIG. 9-3). Then, the INDEX1 (FIG. 9-7) is created with reference to the dictionary 1 (FIG. 9-5) and the dictionary 2 (FIG. 9-6).

On the other hand, a document management application that operates in conjunction with an accounting system extracts the contents (accounting-specific input results) entered in a document file (image) using multiple input methods such as query formula input, and lists them as comments. INDEX2 (FIG. 11) is created with reference to the accounting dictionary 1 (FIG. 9-9) dictionary 2 (FIG. 9-10).

That is, in the conventional document management application, as shown in INDEX1 (FIG. 9-7), the accounting process is not always accurately grasped. Although a document having necessary information could not be specified and searched, a document management application that operates in cooperation with an accounting system can search for an exact match using a plurality of accounting keywords as shown in INDEX2 (FIG. 11). Therefore, it is possible to search for a document suitable for the accounting process. This time, the search process using accounting keywords has been explained. However, the search process using free keywords can be performed with the same mechanism by preparing a free keyword table (FIG. 9-3, etc.) and a dictionary (FIGS. 9-5, 9-6). Can be done in the same way.

FIG. 6-4 shows a processing screen from S3-2-3 to S3-2-6.
The search for document images (certificate images) selects the date, amount, subject, and other keywords (explanation, type of transaction, etc.) as the element items of interest, and based on the range conditions corresponding to the elements of interest. Develop search conditions. It should be noted that the point of designating the element of interest is the same when the vicinity of the image item is clicked or in the case of FIG.

Search conditions (S3-2-6) are date (enter start date and end date when specifying period), amount (enter upper limit and lower limit when specifying range), subject (pull-down method) Select desired subjects, travel expenses, transportation expenses, salary, etc.), free keywords, etc. For the subject and the free keyword, a plurality of items can be input as an OR condition if they are on the same line, and a search can be performed between each item line as an AND condition. The display method (S3-2-5) specifies the display format of the retrieved result (image), and the display format when displaying the image in the book format is the ledger format, trial balance table format, journal book format. Select from any of these.

After entering the search conditions and display conditions, click the [Search] button to execute the search and move to the next process. [Clear] is a button for clearing the input value when the input search condition or display condition is wrong. Note that the subjects are displayed in descending order and ascending order in FIG. 6-4, but may be displayed in their own subjects and other subjects, or in debit and credit subjects.

Next, based on the search condition input in S3-2-6, the search unit searches the image DB in the storage unit. (S3-2-7) In the search, the search index (accounting keyword) given to the image is searched for text. At this time, the weighting information of the search index described above is used for the search.

For example, when a plurality of search condition free keywords are input under OR conditions, weighting information is used when sorting search results. If the corresponding test image does not hit in the course search, the course is set to the counterpart course in FIG. 6-4 and the search is performed again. The counterpart item, which is additional information (not shown) of the accounting keyword used in the previous search, is read out, and the read-out counterpart subject is set as a new accounting keyword and is searched again.

As additional information, accounting-specific information such as the counterpart subject, date, and amount is described, and it is possible to search again not only for the subject but also for the date and amount. If the search is not found in the re-search, the search using the accounting rule may be switched to the search using the free keyword.

The retrieved results are generated in a list format (not shown). (S3-2-8) Based on the generated list, image data and keywords are read from the image DB and displayed on the screen according to the display format set in S3-2-5. (S3-2-9)
In S3-2-9, if there is no image data, a message of “No image” is displayed.

<List of search results>
Although the processing flow has been mainly described above for the search processing, enumerated display of document images (certificate image) obtained by searching will be described.

Fig. 7-1A shows (1) Question type input (documents such as contracts), (2) Question type input (vouchers such as receipts), and (3) Journal entry while looking at images such as receipts. The image data corresponding to the keyword generated from is displayed in the specified display format (in this example, the ledger format is used as an example), and the keyword is formatted at the top of each image data It is displayed.

The screen 800 is explained as follows: The column with keywords of 7/2 rent, withdrawal, 10,800 yen and the image of the rental contract immediately below are as follows: (1) Question type input (documents such as contracts ) Is the image data corresponding to the keyword generated from.

7/3 Welfare expenses, withdrawals, 1,200 yen keyword column and the receipt Nine TeN image below it are generated from (3) journal entries while looking at the receipt image The image data corresponding to the selected keyword.

7/5 Benefits, withdrawals, columns with keywords of 2,500 yen and the image of the receipt family drug immediately below are generated from (2) question type input (certificates such as receipts) The image data corresponding to the keyword, and July 5, 2015 and ¥ 2,500 in the image are surrounded by a frame line, and the enclosed part is recognized by the OCR process and data is acquired. It is shown that it is a place.

7/7 subject unknown, unknown, SKT-48 / 5 pcs keyword field and the image of the invoice immediately below are keywords generated from journal entry while looking at images of receipts, etc. (3) The corresponding image data indicates that “Unknown subject” or “Unknown” is displayed as a keyword because the journal is not completed.

In addition, as shown in Fig. 7-1B, the column with keywords for 7/2 rent, withdrawal, 10,800 yen and the image of the rental contract immediately below are as follows: Since it is image data corresponding to the keyword generated from the document such as a document), a pop-up screen (810) is provided so that it can be understood how and what question is answered when the image of the rental contract is clicked. ) Is displayed.

FIG. 7-2A shows (3) image data corresponding to a keyword generated from a journal entry while viewing an image such as a receipt in a designated display format (here, a ledger format is taken as an example). Shows the displayed screen. If there is no image data, the message “No image” will be displayed instead of the image data. May be displayed.

The case where there is no document image means, for example, when the user inputs while looking at the actual passbooks such as a bankbook, and the search target is not a comment or keyword added to the document image, but a later-described second embodiment. In this case, it is assumed that the display screen is configured when the journal data is included as a search target.

The user can perform a related search (S3-2-10) for searching for a more related image from the displayed images and comment data. S3-2-10 is a subroutine, and the detailed flow is shown in FIGS. 8-1 and 8-2.
FIG. 8A corresponds to the processing when the query formula input (documents such as a contract) S3-1-8 and the query formula input (certificates such as a receipt) S3-1-9 are performed. 8-2 corresponds to the processing when the journal entry S3-1-10 is viewed while viewing the receipt screen.

<Details of related document image (certification image) search processing>
From here, the related search will be described first with reference to FIGS. 8-1, 7-1A, 7-1B, 7-1C, 7-1D, and 7-2B.

FIG. 7-1A shows a screen in which image data corresponding to a keyword is displayed in a specified display format (in this example, a ledger format) in order of date. From the list, select an image or keyword that will be the basis for the related search. (S8-1-1).

FIG. 7-1C shows a display screen when the image data corresponding to the keyword is a trial calculation table format designated as the display format. This shows a state in which image data corresponding to the keyword to be aligned is displayed.

According to the figure, the document images (certificate images) are listed in order of the welfare expense items, and the order of the items (predetermined account items) according to the item structure of the trial balance sheet is shown. It can be seen that they are listed and displayed in a form corresponding to the set order (ascending order, descending order, etc.).

There are the following two methods for instructing the selection of the image or keyword display portion as the reference for the related search.

A) Left-click with the mouse pointer in the vicinity of the keyword display area, and specify the keyword as the search condition. (S8-1-2, S8-1-3)

B) Right-click to call up a search selection menu from a search related information temporary storage area, which will be described later, and display it as a pop-up menu to set search conditions. (S8-1-2, S8-1-4, S8-1-5)

FIG. 7-2B shows a screen for setting a search condition by right-clicking to display a pop-up menu (840). The sorting method can be selected from a trial calculation table format, a ledger format, or a journal format. As a detailed setting, any one of a subject, a date, an amount of money, and a description can be selected. Detailed settings can be set in ascending order and descending order as search conditions.

Information necessary for setting a search condition is called from a search related information temporary storage area to be described later, and when the search condition is set (S8-1-6) by either method, the search related information is temporarily stored. After the search conditions set in the storage area are registered, the image DB is searched based on the registered search conditions. (S8-1-7)

In the case of a), for example, if the subject portion of the keyword display is designated with a mouse pointer or the like and the search is executed by left-clicking, the image data having the same subject as the keyword is relatedly searched.

In the case of (b), referring to FIG. 7-2B as an example, a pop-up menu is displayed by specifying image data with a mouse pointer or the like and right-clicking.

As for the contents of the menu, an item related to the display format and a display format for displaying an image in the book format are selected and instructed from any of the ledger format, trial balance table format, and journal book format. Items for detailed settings and items for which related search should be focused are set. One or more items can be selected from subjects, dates, amounts, and descriptions. It is also possible to set the sort order, ascending or descending order.

In FIG. 7-2B, the pop-up menu is displayed in the trial balance format, narrowed down by subject, and sorted in ascending order, and the related search is executed by pressing the [OK] button in the pop-up menu. . [Cancel] is a button for clearing the input value when the input search condition or display condition is wrong. In addition, unlike a normal search, as a search dedicated to accounting processing, a subject may be searched by switching a credit item and a debit item.

The index of the accounting keyword that matches the search condition from the accounting index registration area of the search index registration area is registered in the search result registration area for accounting processing in the search related information temporary storage area as a result of searching from the image DB. Then, an image corresponding to the keyword as the search result is displayed on the screen (S8-1-8).

FIG. 7-1D shows a screen on which the results of the additional search specified in FIG. 7-2B are displayed as a list. As another example, data in which image data, comment data, and a search index (accounting keyword search, free keyword search) are integrated may be registered in the image DB.

Here, the display formats of the ledger format and the trial calculation table format will be described using the tables of FIGS. 5-12 to 5-15. Each of the tables in FIGS. 5-12 to 5-15 has a structure in which data is stored in units of items such as codes, classifications, and subject names. The relationships between the tables are based on a common code. Based on this code, the display format of the ledger format and trial calculation table format is determined. With reference to the relations in FIGS. 5-12 to 5-15, the aggregation and rearrangement of the image data are processed based on the keywords corresponding to the image data.

In displaying keywords, when the image database is searched at the time of search execution, if there are enough keywords for display in the image data to be searched, the keywords are displayed with priority. If there are not enough keywords, The comment may be displayed as it is instead of the keyword, or the part lacking the keyword may be supplemented and displayed.

When the display of all searched images is completed (S8-1-12: Y), the state transitions to the next instruction wait state.

When the user does not find a reference test image or when he / she wants to set a new search condition and perform a search, he / she presses the ESC key from the keyboard, or in FIG. Right-click a location where no data is displayed, and select and instruct re-search (not shown) from the menu. The processing corresponds to whether the search is still performed (S3-2-11: Y), and another search can be performed by returning to S3-2-3. When the search is not performed (S3-2-11: N), the process is terminated.

Further, the related search will be described with reference to FIGS. 8-2, 7-2A, 7-2B, and 7-1D. FIG. 7-2A shows a screen in which image data corresponding to a keyword is displayed in a specified display format (here, the ledger format is taken as an example) in order of date. From the list, select an image or keyword that will be the basis for the related search. (S8-2-1).

Since the processing of S8-2-2 to S8-2-7 is the same processing as S8-1-2 to S8-2-7 described above, description thereof will be omitted.

In S8-2-8, a determination is made as to whether or not there is an image serving as a reference for the related search. FIG. 8-2 corresponds to the process when the journal entry S3-1-10 is viewed while viewing the receipt screen, but the journal entry does not necessarily have an image. For example, the bankbook If you enter while looking at the actual passbooks, the image will not exist. Taking FIG. 7-2A as an example, there is no image of the journal entry for 7/2 deposit 150,000.

If there is an image (S8-2-8: Y), the process proceeds to S8-2-10, and an image corresponding to the keyword is displayed on the screen. If there is no image (S8-2-8: N), the process proceeds to S8-2-9, and the message “No image” and the keyword are displayed on the screen. In displaying keywords, when the image database is searched at the time of search execution, if there are enough keywords for display in the image data to be searched, the keywords are displayed with priority. If there are not enough keywords, The comment may be displayed as it is instead of the keyword, or the part lacking the keyword may be supplemented and displayed.

When the display of all the searched images is completed (S8-2-11: Y), the state transitions to the next instruction wait state.

<Details of document image information (comment etc.) input processing>
From here, corresponding processing will be described using FIGS. 4-1, 4-2, and 4-3. Each figure corresponds as follows.
Fig. 4-1 Detail of Question Formula Input (Documents such as Contract) S3-1-8 Fig. 4-2 Detail of Query Formula Input (Voucher such as Receipt) S3-1-9 4-3 : Journal entry while looking at the receipt screen Details of S3-1-10

First, with reference to FIG. 3A as appropriate mainly with reference to FIG. 4-1, the question type input (documents such as contracts) will be described. In S4-1-1, it is determined whether recognition data has been acquired for the result of the OCR process in S3-1-5 in FIG. When there is recognition data (S4-1-1: Y), among the recognition data, the date and amount, which are accounting elements, are temporarily stored as the answer to the question formula, and the corresponding date, amount column, etc. on the question formula input screen Recognition data is set in the location (S4-1-2), and the process proceeds to S4-1-3. If there is no recognition data (S4-1-1: N), nothing is set in the answer, and the process proceeds to S4-1-3.

The user inputs an answer based on a description rule (specific rule) specific to the accounting process or tax process while referring to an image such as a contract. (S4-1-3) For input, an answer is selected from a plurality of choices or a text input.

Answer registration data is generated from the answer input data. (S4-1-4) The generated data is output as a comment. (S4-1-5)

Next, summary registration data is generated from the input data. (S4-1-6) The summary registration data generated is an answer to a question, with characters such as “/” as a delimiter so that the user can understand the contents of the transaction or contract. FIG. 6-2 For example, the character string in the summary column displayed at the bottom of FIG. 6-2 corresponds to “item purchase less than 100,000 / adhesive plaster / tea / ordinary medicine / uniform / etc. / drug store Sato ///”.

The generated summary registration data is also output as a comment. (S4-1-7)

Next, in order to process whether a subject can be generated from an answer, a table shown in FIG. 5-11 for question answer data linked subject generation is read from the storage unit. (S4-1-8). FIG. 5-11 shows the conditions for generating subjects from the summary in a table format. The items in the table include a code, a debit account item, a credit account item, and a description character string.

The read table and the answer data for the question are checked. (S4-1-9) In the check, the subject is generated by collating the summary character string in the table of FIG. 5-11 in the text data of the summary registration data generated in S4-1-6. In the example here, the words of tea and regular medicine are collated from the character string in the above description column, "item purchase less than 100,000 / adhesive plaster / tea / ordinary medicine / uniform etc./drug store Sato ///" Judgment can be made as welfare expenses for debit items and cash for credit items.

If the subject can be generated (S4-1-10: Y), the subject registration data is generated with reference to the table shown in FIG. 5-11. (S4-1-11) The generated subject is output as a comment, and the process proceeds to S4-1-13.

When the course cannot be generated (S4-1-10: N), the process proceeds to S4-1-13. In S4-1-13, the text entered in the remarks column and the like input by the user in addition to the answer to the question is output to the comment.

4-2 Questionnaire Input (Voucher such as Receipt) Details of S3-1-9, S4-2-1 to S4-2-13 are S4-1-1 to S4-1 in FIG. Since it is the same process as −13, the description is omitted.

4-3 will be described. FIG. 4-3 shows the details of journal entry S3-1-10 while viewing the receipt screen. FIG. 4A and FIG. 4B are processings when answering and inputting a question, but differ in that journal entry is performed. In S4-3-1, it is determined whether recognition data has been acquired for the result of the OCR process in S3-1-5 in FIG.

When there is recognition data (S4-3-1: Y), the date and the amount of money are registered in the journal data (S4-3-2), and the process proceeds to S4-3-3. If there is no recognition data (S4-3-1: N), nothing is set in the journal data and the process proceeds to S4-3-3.

While referring to an image such as a receipt, the user inputs journals such as subjects and descriptions to generate journal data. (S4-3-4) The generated data is output as a comment. (S4-3-5)

Summary registration data is generated from the summary entered as part of the journal. (S4-3-7), the generated summary registration data is also output as a comment. (S4-3-8)

Course registration data is generated from the course entered as part of the journal. (S4-3-9), the generated subject registration data is also output as a comment. (S4-3-10)

<Function block configuration>
The hardware configuration of the data management system of the present invention will be described here. FIG. 1 is a diagram showing the hardware configuration of an accounting processing apparatus (consultant's terminal) 10 in units of blocks. An input unit 11, an output unit 12, a control unit 13, a storage unit 14, and a communication unit are shown in large units. 15 and an image capturing unit 16.

The input unit 11 is means for transmitting a primary operation from the user to the control unit, and generally corresponds to a keyboard, a mouse, or the like, and may be a touch panel or the like. Also, a microphone for voice input may be used. The image capturing unit 16 is a means for imaging a receipt, a contract, or the like, and is generally a scanner or a digital camera, and may be a smartphone or a tablet as long as it has an imaging function. . The output unit 12 is output means such as a display (LCD, CRT, etc.) and printing (printer).

  The control unit 13 is a functional unit that performs control in the terminal device by a program loaded in the CPU and the storage unit 14, and allows the operator to select document data from the input unit 11 and to perform user operations such as data registration processing instructions. Selection and processing instruction unit to be accepted, data management unit for reading and saving files, taking in and converting data based on user operations, etc., transaction generation unit for reflecting transaction data in journal data, An OCR processing unit that generates text data by performing OCR processing on the captured image, a keyword generation unit that generates a keyword that serves as a search index, and a predetermined table or dictionary (table storage area of the storage unit 14). Search text information of extracted document data, extract keywords, search DB in storage Text search unit that refers to the table or dictionary based on the extracted keyword has the storage destination determination unit for determining a storage destination or a candidate of the document data. The function (function) of each part will be described later.

The storage unit 14 corresponds to a storage device such as a hard disk or an SSD, and includes an accounting processing storage area 141, a shared processing storage area 142, and a filing processing storage area 143. The accounting processing storage area 141 includes a question answer DB (not shown) and a journal data DB (not shown). Both processing shared storage areas 142 include a program storage unit (not shown), an input data DB (not shown), a question DB (not shown), a transaction information temporary storage area (not shown), and an image DB (not shown). ).

The filing processing storage area 143 includes a search related information temporary storage area (not shown), a document data group storage area (not shown), a file storage unit (not shown), and a table storage area (not shown). Shared storage area 1431, dictionary DB (not shown), free keyword search (storage area) 1432 composed of text index registration area (not shown), accounting dictionary DB (not shown), accounting index registration The account keyword search (storage area) 1433 is composed of an area (not shown).

The program storage unit and image DB storing various application programs store image images such as receipts and receipts. Question DB is 5W1H (when, where, who, what, why, why) or 5W2H (when, where, who, what) based on accounting rules or specific rules for accounting (specific rules) , How, why, how much) DB that stores questions, input data DB by OCR processing, question formula input (documents such as contracts), question formula input (certificates such as receipts) and manual input It is a DB for storing input data input by a plurality of input methods such as journal entry (accounting input).

An OCR dictionary DB (not shown) is a DB that stores a dictionary that is referred to during OCR processing. The temporary storage area for accounting information is a storage area for registering subjects linked with question type input (documents such as contracts) responses and question type input (certificates such as receipts) responses.

The journal data DB transmits the input data of manual journal input to the accounting system, and holds information on past journals for which accounting processing has been performed, for example, as monthly journal data. Since the monthly journal data is created according to the closing date, for example, if the closing date is 20, the monthly journal data for November is for transactions that occurred on (October 21 to November 20). Retain data. The data management unit can manage a plurality of corporations and businesses. Therefore, the image DB and journal data DB described above manage data on a company basis. Files, data, and tables stored in the storage unit will be described later.

Question type answer DB transmits the answer (input data) input by question type input (documents such as contracts) and question type input (certificates such as receipts) to the accounting system, and then performs accounting processing. This is a DB for storing answers that can be used in the accounting process.

The shared storage area 1431 includes search condition management information, document management display screen configuration information, search selection menu registration information, and search processing information temporary storage area for temporarily storing registration information for accounting processing search conditions. A document data group primary storage area for temporarily storing document data whose registration location is to be determined, a table storage area for registering various tables or dictionaries necessary for determining the registration location of document data, and finally This is a storage area that is configured by a file storage unit in which document data is stored and is shared by both free keyword search and accounting keyword search.

The free keyword search (storage area) 1432 is a storage area in which storage areas used for free keyword search are collected.
The dictionary DB is a DB storing keywords and the like used for document management, and is referred to for weighting of a search index.
The text index registration area is a storage area for generating and registering an index used for free keyword search from a dictionary storage area and a table storage area storing comments.

The accounting keyword search (storage area) 1433 is a storage area in which storage areas used for accounting keyword search are collected.
The accounting dictionary DB is a DB that stores keywords and the like used for accounting processing, and is referred to for weighting of a search index.
The accounting index registration area is a storage area for generating and registering an index used for accounting keyword search from the accounting dictionary DB and a table storage area storing comments.

Although the accounting processing apparatus (consultant terminal) is a stand-alone PC, a cloud terminal may be used to provide a storage unit in an external storage (server) and use a tablet terminal or the like as a thin client.

The communication unit 15 corresponds to a LAN adapter or the like, and has a function for connecting to a local network-connected server or the Internet.

FIG. 2A shows a client-server configuration which is a basic configuration, and all of the above-described embodiments are described with this configuration in mind. As shown in the figure, the terminal device 10 or 30 of the consultant is a client terminal, and the client terminal 10 is connected to the server device 20 installed in the accounting office via the network. The client terminal device 30 is operating. In this example, necessary data may be stored in the server device 20 of the accounting office, and in this case, a public cloud, a hybrid cloud, and a private cloud are used. Further, the client terminals 10 and 30 may be home input terminals or remote terminals.

In the embodiment shown in FIG. 2B, the server device 20 on the accounting office side provides services to the consultant in the form of a virtual machine. That is, the computer resource of the server device 20 is allocated to each consultant as a virtual machine, and the consultant terminal device 10 operates as an input / output device. The terminal device of the advisor is described as a personal computer arranged in the advisor company in this embodiment, but the terminal device is a smartphone or tablet PC as long as it can provide various functions and data of the system described below. Further, various functions of the terminal device can be realized as a web service by a web server accessible from the PC of the consulting company. Each means, function, and method are not limited to the configurations in FIGS. 2-1 and 2-2, and may be in either the server or the client terminal.

In FIG. 2-3, the storage unit of the server device 20 on the accounting office side is installed in a data center or the like, and the terminal device 10 of the consultant or the client terminal device 30 of the accounting office accesses necessary data via the network. A system for providing services on the cloud is configured as a cloud server form.

FIG. 2-4 is an embodiment in which the client terminal device 30 of the accounting office in addition to the consulting terminal device 10 is provided with services in the form of a virtual machine by the server device 20 installed in a data center or the like. Thus, the bookkeeping system or accounting system of the present invention can be realized in various forms.

As mentioned above, although the aspect which mainly gives a comment and a keyword to a document image has been demonstrated, it is not restricted to this, For example, as input data other than a document image, input text data (comment, keyword, book table text data, In a system that stores data (journal data) and manages association with document images, the data management system related to test image search and enumeration according to the present invention can be applied.
More specifically, accounting-specific keywords are extracted from the journal data that has been input using the accounting dictionary 2 for extracting keywords as the search index, and the index is added to the test image. After the addition, the search process of the first embodiment may be executed (... an eleventh invention).
Alternatively, it is possible to execute the search process of the first embodiment by directly using each item element of journal data that has been input, for example, date, amount, description, and subject (Twelfth Invention). ).

  The eleventh invention and the twelfth invention will be described below.

In the eleventh aspect of the present invention, the accounting input that has been registered in the accounting processing storage area (141 in FIG. 1) is linked with the entered image data and the journal data that has been used in the accounting input system. In the system, accounting keywords are extracted from journal data, recorded as accounting indexes of the data management system, and images are listed and displayed according to the specified book format (ledger format, trial balance format, journal format, etc.) It is.
Fig. 3-2. This is the process of S3-2-1 to S3-2-2 in the overall flow.

At this time, how to extract a keyword as a search index from journal data will be described.
Journal data generated by inputting transaction information from the accounting input application is shown in Fig. 5-7. It is registered in the journal data DB in a format like a journal data management table.
This process is a normal journal data generation process, which is shown in FIG. The entire flow does not pass and is generated by the accounting process.
The journal data is registered in the journal data DB in the storage area of the server device. In the case of a cloud-based data management system, the journal data is registered in the journal data DB in the storage area on the cloud. However, it may be registered in a storage area of another device according to the system configuration. On the other hand, the captured image data is shown in FIG. 5-1. It is registered in the image DB in the image management table.

Here, character information is extracted from the journal data in order to search for and enumerate the testimony images. Register in the accounting input item storage area in the input data DB of the storage unit described in the overall flow.
In the accounting input item storage area, FIG. Keyword information management table, FIG. Accounting keyword information management table, Fig. 5-4. A free keyword information management table is included.
In addition, FIG. The image management table is included in the image DB.
In order to acquire character information from the journal data as a comment, extract a keyword from the character information, and assign it to the test image data as a search keyword, first, character information is acquired from the journal data, and FIG. It is registered in the keyword information management table (S3-2-1).

Data of each item of date, amount, and summary registered in the journal data management table (FIGS. 5-7) is shown in FIG. 5-3. Register in the corresponding item of the accounting keyword information management table.
The items described in the accounting keyword information management table (FIG. 5-3) are merely examples, and other items (consultant, subject) may be additionally registered.
In addition, information such as date, amount, and business partner from journal data is shown in Fig. 5-4. It is also automatically registered in the free keyword information management table. The items described in the free keyword information management table (FIG. 5-4) are merely examples, and other items (other character information in the journal data) may be additionally registered.
The information in the accounting keyword information management table (FIG. 5-3) and the information described in the free keyword information management table (FIG. 5-4) are shown in FIG. Register in association with the keyword information management table.
Up to here, FIG. 3-2. This is the process of step S3-2-1 in the overall flow.
In this embodiment, the table is divided into a plurality of tables, but may be configured as one table as shown in FIG. 9-8.

In the image management table (FIG. 5-1), text information (character information) about the testimonial image is stored as a comment. Therefore, a keyword as an index for search processing is extracted from the comment, and the testimonial image search, 9-11 used in enumeration. By registering in INDEX2 (accounting keyword), modification information such as comments and keywords is assigned to the test image.
The process of extracting a keyword from character information (comments) is shown in FIG. Accounting dictionary 1 (accounting keyword for accounting) and Fig. 9-10. A keyword is extracted from character information (comments) in consideration of weighting as an accounting keyword based on the accounting dictionary 2 (accounting keyword) (S3-2-2). When using as an accounting keyword, an accounting keyword flag is set.
Although the accounting keyword flag is used in the present embodiment, an accounting keyword index (FIGS. 9-11) may be prepared separately. As described above, the journal data is decomposed into a system that manages the test image and optimized as an accounting keyword, thereby making it easy to use the test image search and enumeration on the data management system.
Up to here, FIG. 3-2. This is the process of step S3-2-2 in the overall process flow.

In the twelfth invention, in the system in which the test image and the accounting data input with respect to the test image are linked and recorded in the DB, the data management system uses the accounting data as they are and relates other related test images. The image is enumerated and displayed according to the specified book format (ledger format, trial balance format, journal format, etc.).
Fig. 3-2. The processing of S3-2-3 to S3-2-9 in the entire flow is applicable.

  When searching for and enumerating testimony images, INDEX2 (accounting keywords) (Fig. 9-11) and image + comment + keyword table (Fig. 5-8) managed by the data management system are used. If the relevant image is not sufficiently generated, is not generated, or does not exist, the data management system uses the image that is managed by the accounting input system. Since it is also possible to search and enumerate the proof image, it is possible to search and enumerate the proof image in an integrated manner.

When journal data is used as a direct accounting keyword in the search process (Fig. 3-2. S3-2-7 in the overall flow), in the test image search, the text information in the journal data (Fig. 5-7) is accounted for. Based on the weighting in dictionary 2 (accounting keyword) and the priority set in the receipt image search range condition table (Fig. 5-5), test image search and enumeration are performed ( 3-2, S3-2-8 to S3-2-9 of the overall flow).

The characteristics of the present invention will be described together.
Feature 1 of the invention is that, for each item set based on the accounting rules, it accepts the input of keywords and indexes for accounting processing relating to the testimony image, adds them to the image, and designates the specified book format (ledger format , Trial image list format, journal book format, etc.).

Characteristic point 2 of the invention is that image modification information is given by inputting a query for a testimony image.

The feature point 3 of the invention is that image modification information is given by accounting input for the test image.

The feature point 4 of the invention is that the image modification information is given by the text information acquired by the OCR preprocessing for the test image.

The feature point 5 of the invention is to search and enumerate related testimony images based on the target element and the range condition in accordance with the accounting rule for each element.

The feature point 6 of the invention is automatic search and manual selection of the element of interest.

The feature 7 of the invention is that journal data is created by combining image modification information of the test image.

The feature 8 of the invention is that monthly and annual audits are performed by referring to receipts between books that are endlessly vertical and horizontal.

The feature point 9 of the invention is that in a system in which the test image and the accounting data inputted with respect to the test image are linked and recorded in the storage area, elements constituting the accounting data are extracted and the image modification is performed on the test image. It is given as information and enumerates and displays images according to the specified book format (ledger format, trial balance format, journal format, etc.).

The feature 10 of the invention is that in the system in which the test image and the accounting data input with respect to the test image are linked and recorded in the storage area, other related test images are searched using the accounting data as they are, The list of images is listed and displayed according to the specified book format (ledger format, trial balance format, journal format, etc.).

Although the embodiments of the present invention have been described in detail above, the technical scope of the present invention is not limited to the above-described embodiments or examples, and the present invention does not depart from the scope of the appended claims. Various modifications and modifications can be realized. All such modifications and changes should be understood as belonging to the technical scope of the present invention.

The accounting method according to the present invention presents the user with a comment that summarizes the journal data and image images such as documents (contracts, receipts, receipts, etc.) that originated in the form of a book. The user can intuitively verify the journal data, and it can be used as a reference in transaction input, so that even those who have little knowledge of bookkeeping can easily verify and input the journal data.

10 Accounting processing device (consultant terminal)
20 servers (accounting office side)
11 Input unit 12 Output unit 13 Control unit 14 Storage unit 15 Communication unit 16 Image capturing unit

Claims (17)

A format that accepts the input of text information for accounting processing, adds it to the image as image modification information, searches for the image using the image modification information, and sets it based on accounting rules A data management system for enumerating testimony images along
From the storage area where the testimonial image is stored, obtain and present a specific testimonial image,
Accept or get input for each element set based on accounting rules for the displayed proof image
Image modification information recording means for recording the input or acquired text information as image modification information for modifying the display image;
Search means for searching for a testimony image using image modification information;
Specify whether to enumerate on the basis of a specified book format including at least one of the ledger format, trial balance format, and journal format as the format for enumerating the testimony images obtained by searching in order to,
Image enumeration style designation means,
Listing the testimony images obtained from the search according to the selected book format;
A data management system characterized by
As the text information when inputting or obtaining for each element set based on the accounting rule, the above-mentioned testimony image,
Each element has a question type input acceptance means that presents a set of questions and answer options,
A set of questions and selected answer options
An image modification information recording means based on a query expression that is added and recorded as image modification information for modifying a testimony image,
The data management system according to claim 1, further comprising:
As text information when inputting or obtaining for each element set based on the accounting rule for the above-mentioned proof image,
For each book and slip element, there is an accounting data input receiving means for receiving data input,
An accounting data input-based image modification information recording means that records input data for each received element as image modification information for modifying the display image,
The data management system according to claim 1, further comprising:
As the text information when inputting or obtaining for each element set based on the accounting rule, the above-mentioned testimony image,
A text information recognition acquisition means for acquiring meaningful text information on accounting rules by performing character recognition pre-processing on the testimony image,
A character recognition pre-processing based image modification information recording means for adding and recording the acquired text information as image modification information for modifying the display image,
The data management system according to claim 1, further comprising:
A data management system for searching and enumerating other testimony images related to a given testimonial image,
When a specific ID image is selected from multiple ID images,
As a means of searching for and listing other proof images related to the selected proof image,
An element of interest designating means for designating which element of the image modification information given to the specific testimonial image is to be focused on;
For each element of interest, a search range condition including the same month in the previous year or amount range based on accounting rules or a search range condition including switching the search target from the subject to the other subject is stored as accounting-specific processing. Search range condition storage means for storing in the section;
Based on the specified element of interest and the search range condition corresponding to this element of interest,
Means for searching for image modification information recording means,
Searching for and listing other proof images associated with the identified proof image;
The data management system according to any one of claims 1 to 4.
As the element of interest designating means,
Means for automatically acquiring as a focused element by instructing a predetermined item or neighborhood of the testimony image, or
Means for separately displaying and selecting a list of element items,
This element of interest designation means,
For each element of interest, the search range condition including the same month in the previous year and the amount range based on the accounting rule, or the search range condition including switching the search target from the subject to the counterpart as a process specific to the accounting is set. Search range condition storage means for storing in the storage unit,
On the basis of the,
Searching for and listing other proof images associated with the identified proof image;
The data management system according to claim 1, wherein:
Create journal data by combining the image modification information of the proof image obtained by searching,
The data management system according to claim 1, wherein:
While switching each book, based on the proof image obtained by searching or its image modification information,
A means to confirm the appropriateness of accounting in each book;
The data management system according to claim 1, wherein:
A format that accepts the input of text information for accounting processing, adds it to the image as image modification information, searches for the image using the image modification information, and sets it based on accounting rules And a data management method for enumerating testimony images,
From the storage area where the testimonial image is stored, obtain and present a specific testimonial image,
Accept or get input for each element set based on accounting rules for the displayed proof image
An image modification information recording step for recording the input or acquired text information as image modification information for modifying the display image;
A search step for searching for a testimony image using image modification information;
Specify whether to enumerate on the basis of a specified book format including at least one of the ledger format, trial balance format, and journal format as the format for enumerating the testimony images obtained by searching in order to,
An image enumeration style specification step,
Listing the testimony images obtained from the search according to the selected book format (in the order of the items in the book);
A data management method characterized by the above.
A data management method for searching for and enumerating other proof images related to a predetermined proof image,
When a specific ID image is selected from multiple ID images,
As a process of searching for and listing other testimonial images related to the selected specific testimonial image,
An element of interest designating step for designating which element of image modification information given to the specific testimonial image is to be focused on;
For each element of interest, set a search range condition based on accounting rules, including the same month in the previous year, including the amount range, or a search range condition as a process specific to accounting (including switching the search target from the subject to the other subject) A search range condition storage step stored in the storage unit;
Based on the specified element of interest and the search range condition corresponding to this element of interest,
Searching for image modification information recording means,
Searching for and listing other proof images associated with the identified proof image;
The data management method according to claim 9.
In the system in which the accounting data entered for the proof image is associated with the proof image and recorded in the storage area,
A form in which each element of the accounting data associated with the testimony image is extracted, added to the image as image modification information, and the testimonial image is searched using the image modification information and set based on the accounting rules A data management system for enumerating testimony images along
Extract the elements that make up each accounting data,
For each file, a testimonial image file associated with the accounting data
Image modification information providing means derived from accounting data to be provided as image modification information for modifying the testimony image;
Search means for searching for a testimony image using image modification information;
Specify whether to enumerate on the basis of a specified book format including at least one of the ledger format, trial balance format, and journal format as the format for enumerating the testimony images obtained by searching in order to,
Image enumeration style designation means,
Listing the testimony images obtained from the search according to the selected book format;
A data management system characterized by
In the system in which the accounting data entered for the proof image is associated with the proof image and recorded in the storage area,
This is a data management system that uses the elements that make up accounting data to search for other related images, and enumerates other related images according to the format set based on the accounting rules. And
Using each element that makes up accounting data,
A search means for searching for other proof images related to a given proof image;
Enumerate based on a specified book format including at least one of the ledger format, trial balance format, and journal format as a format for enumerating other related testimonial images obtained by searching To specify whether to
Image enumeration style designation means,
Listing the other relevant evidence images obtained from the search according to the selected book format;
A data management system characterized by
A data management system for searching and enumerating other testimony images related to a given testimonial image,
When a specific ID image is selected from multiple ID images,
As a means of searching for and listing other proof images related to the selected proof image,
An element of interest designating means for designating which element of the image modification information attached to the specific testimonial image or the accounting data associated with the testimonial image is to be focused on;
For each element of interest, search range conditions including the same month in the previous year and amount range based on accounting rules, or search range conditions including switching the search target from the subject to the other subject as a process specific to accounting are stored and stored. Search range condition storage means for storing in the section;
Based on the specified element of interest and the search range condition corresponding to this element of interest,
Image modification information recording means or means for searching each element constituting accounting data,
Searching for and listing other proof images associated with the identified proof image;
The data management system according to any one of claims 11 to 12.
As the element of interest designating means,
Means for automatically acquiring as a focused element by instructing a predetermined item or neighborhood of the testimony image, or
Means for separately displaying and selecting a list of element items,
This element of interest designation means,
For each element of interest, the search range condition including the same month in the previous year and the amount range based on the accounting rule, or the search range condition including switching the search target from the subject to the counterpart as a process specific to the accounting is set. Search range condition storage means for storing in the storage unit,
On the basis of the,
Searching for and listing other proof images associated with the identified proof image;
The data management system according to any one of claims 11 to 13, characterized by:
Creating journal data by combining the image modification information of the certificate image obtained by searching or the elements of accounting data associated with the certificate image;
The data management system according to claim 11, characterized in that:
Testimonial image obtained by searching while switching each book, or
Based on the image modification information or the accounting data associated with the proof image,
A means to confirm the appropriateness of accounting in each book;
The data management system according to any one of claims 11 to 15, characterized by:
The data management program for functioning a computer system as each means in the data management system as described in any one of Claims 1-8 or Claims 11-16.

Images