JP7549842B2 - Automatic sorting device, control method for automatic sorting device, and control program for automatic sorting device - Google Patents

Description

The present invention relates to an automatic sorting device for automatically sorting documents, a control method for the automatic sorting device, and a control program for the automatic sorting device.

Conventionally, there has been a proposal to input information on receipts, etc., using an optical character recognition/reader (OCR), and then to journalize information such as the "summary" of the input receipts, etc., select account items, etc., and register them together with information such as the summary, thereby creating a history database.
In the case of such a proposal, when new information such as a receipt is subsequently entered using OCR or the like, the summary and other information is compared with the information in the history database that has already been created, and the most suitable accounting information is selected and provided (for example, Patent Document 1, etc.).

JP 2007-304643 A

However, in such a proposal, it was necessary for the "history database" to have item information that was easy to compare with information such as the "summary" of evidence such as receipts entered using OCR or the like.
In other words, in addition to information such as items for accounting purposes, the history database requires the creation of information that can be compared with information such as specific product information of the vouchers entered, which results in the problem of complex equipment, etc.

The present invention aims to provide an automatic sorting device, a control method for an automatic sorting device, and a control program for an automatic sorting device that can easily and quickly generate accounting information from information contained in input receipts and other supporting documents without complicating databases, etc.

The object of the present invention is to provide a system for providing a document that includes evidence image information, which is image information related to evidence, and pre-registered evidence image information, which is pre-registered evidence image information, and a content information determination unit that compares journalization basic information included in the evidence image information with pre-registered journalization basic information included in the pre- registered evidence image information to generate comparison information, and when the content information determination unit determines that the comparison information matches threshold information, the system selects the pre-registered journalization basic information to be used as usage information for journalization, and when the content information determination unit determines that the comparison information matches threshold information, the system selects the pre-registered journalization basic information to be used as usage information for journalization, and when the content information determination unit determines that the comparison information matches threshold information, the system selects the pre-registered journalization basic information to be used as usage information for journalization, and generates comparison information. This is achieved by an automatic accounting device that extracts already registered journal information associated with the already registered evidence image information of the foundation information, obtains unique information of the evidence image information, and rewrites only the unique information of the already registered journal information to generate journal information of the evidence image information , and has a shape information storage unit that stores information on the color of the previously acquired evidence, the arrangement of items, and the shape and size of the ruled rectangle, wherein the unique information is information on the amount, date, and issuer, and even if the description of the issuer information cannot be obtained, the information on the issuer of the evidence is estimated from the color of the evidence, the arrangement of items, and the shape and size of the ruled rectangle stored in the shape information storage unit .

According to the above configuration, the basic information for journalizing contained in the evidence image information is determined by comparing it with already registered evidence image information, so there is no need to prepare separate information for comparing the basic information for journalizing the evidence image information; it is sufficient to compare the originals, etc. with already registered evidence image information in which they are registered as is, allowing for easy and quick comparison.
Furthermore, new accounting information is generated by changing only the unique information, such as the date and amount, of the already registered accounting information, such as the book-entered data associated with the corresponding already registered evidence image information. Therefore, when generating new accounting information, erroneous input by users, etc. can be prevented, and new accounting information can be generated easily and quickly.
In addition, according to the above configuration, when the comparison information matches the threshold information, a selection is made of already-registered basic information for journalizing to be used as usage information for journalizing, thereby making it possible to generate more accurate journalizing information.
Furthermore, according to the above configuration, the already registered voucher image information includes at least form information (e.g., the color of a receipt or the arrangement of items) including either the shape or color of the voucher, so even if it is not possible to obtain basic information for journal entry (e.g., text information such as the issuer) included in the voucher image information, it is possible to identify the basic information for journal entry from the form information, and new journal entry information can be automatically generated.

Preferably, the pre-registered basic information for journal entry of the automatic journal entry device is characterized in that different weighting information is associated with different types of basic information for journal entry.

According to the above configuration, the weighting information differs depending on the already registered basic information for accounting entries, such as the issuer of the receipt, the title of the magazine, the general name of the product, etc., so that when generating comparison information, it is possible to make a comparison that places emphasis on the basic information for accounting entries with heavier weighting information.
This makes it possible to generate more accurate journal information.

Preferably, the weighting information associated with the pre-registered basic information for journal entry of the automatic journal entry device is automatically changed depending on the processing frequency of the pre-registered basic information for journal entry.

According to the above configuration, the weighting information is automatically changed depending on the processing frequency of each registered basic information for journal entry in the evidence image information, so that the automatic journal entry device can "learn" to automatically generate "comparison information" that is in line with the journal entry processing trends of users such as accounting firms that use the automatic journal entry device. This makes the device easy to use for users.

Preferably, the registered image information of the voucher of the automatic sorting device includes at least morphological information, including either the shape or color of the voucher, and is configured to identify the basic information for sorting based on the morphological information.

Preferably, the accounting information of the evidence image information of the automatic sorting device is the accounting information of an accounting firm.

According to the above configuration, since the accounting information of the document image information is accounting information of the accounting firm, the accounting firm learns the trends of past accounting processing, etc., and the accuracy of the accounting information of the document image information improves with repeated processing.

The above object of the present invention is to provide a method for controlling an automatic sorting device having evidence image information, which is image information related to evidence, and pre-registered evidence image information, which is pre-registered evidence image information, in which a content information determination unit generates comparison information by comparing basic information for journalizing contained in the evidence image information with the pre-registered basic information for journalizing contained in the pre-registered evidence image information, selects the pre-registered basic information for journalizing to be used as information for journalizing based on the comparison information, and extracts the pre-registered journalizing information associated with the pre-registered evidence image information of the selected pre-registered basic information for journalizing. At the same time, unique information which is information on the amount, date and issuer of the evidence image information is obtained, and only the unique information of the already registered accounting information is rewritten to generate accounting information for the evidence image information, and information on the color, item arrangement, and shape and size of the ruled rectangle of the previously obtained evidence is stored in a shape information storage unit, and even if the description of the issuer information cannot be obtained, information on the issuer of the evidence is estimated from the color, item arrangement, and shape and size of the ruled rectangle of the evidence stored in the shape information storage unit.

The above object of the present invention is to provide an automatic sorting device having evidence image information, which is image information related to evidence, and pre-registered evidence image information, which is pre-registered evidence image information, with a function in which a content information determination unit compares journalization basic information contained in the evidence image information with the pre-registered journalization basic information contained in the pre-registered evidence image information to generate comparison information, a function to select the pre-registered journalization basic information to be used as journalization usage information based on the comparison information, a function to extract the pre-registered journalization information associated with the pre-registered evidence image information of the selected pre-registered journalization basic information, and This is achieved by a control program of an automatic accounting device that realizes the following functions: obtaining unique information , which is information on the amount, date, and issuer of the evidence image information, and rewriting only the unique information of the already registered accounting information to generate accounting information for the evidence image information; storing information on the color of the previously obtained evidence, the item arrangement, and the shape and size of the ruled rectangle in a shape information storage unit; and, even if it is not possible to obtain the information on the issuer, estimating information on the issuer of the evidence from the color of the evidence, the item arrangement, and the shape and size of the ruled rectangle stored in the shape information storage unit.

As described above, the present invention has the advantage of being able to provide an automatic sorting device, a control method for an automatic sorting device, and a control program for an automatic sorting device that can easily and quickly generate accounting information from information contained in input receipts and other supporting documents without complicating databases, etc.

1 is a schematic block diagram showing the main configuration of an "automatic journalizing device for evidence 10" according to a first embodiment of the "automatic journalizing device" of the present invention. [0023] FIG. 2 is a schematic block diagram showing the main configuration of a first various information storage unit 20. [0023] FIG. 4 is a schematic block diagram showing the main configuration of a second various information storage unit 40. FIG. 1 is a schematic flowchart showing an example of a main operation of the "voucher automatic sorting device 10" according to the first embodiment. 13 is a schematic flowchart showing the main operations of the automatic document sorting device 10 according to the second embodiment. 11 is a schematic explanatory diagram showing the state of the "first registered receipt image information storage unit (part 1)" at stage ST2. FIG. FIG. 11 is a schematic explanatory diagram showing the state of the "first registered receipt image information storage unit (part 2)" at stage ST7. FIG. 2 is a schematic explanatory diagram showing an example of morphological information stored in a "morphological information storage unit."

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.
It should be noted that the embodiments described below are preferred specific examples of the present invention, and therefore various technically preferable limitations are imposed thereon. However, the scope of the present invention is not limited to these embodiments unless otherwise specified in the following description to the effect that the present invention is limited thereto.

(First embodiment)
FIG. 1 is a schematic block diagram showing the main configuration of an "automatic journalizing device for evidential documents 10" which is a first embodiment of an "automatic journalizing device" of the present invention.
Account information acquisition device 10 in FIG. 1 has a computer, which has a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), a hard disk, etc., and is connected via a bus.

As shown in FIG. 1, the automatic document sorting device 10 has a control unit 11, which controls a "communication device 12" for communicating with external devices, a display unit for displaying various information, such as a "display 13," and various information input devices 14, such as a keyboard, for inputting various information.

As shown in FIG. 1, the control unit 11 also controls a "first various information storage unit 20" and a "second various information storage unit 40", and FIGS. 2 and 3 are schematic block diagrams showing the main configurations of the "first various information storage unit 20" and the "second various information storage unit 40", respectively.
The contents of each block of the "first various information storage unit 20" and the "second various information storage unit 40" will be described later.

4 and 5 are schematic flow charts showing an example of the main operations of the "voucher automatic sorting device 10" according to this embodiment.
In this embodiment, the following description will be given using an example in which an accounting firm, for example, A, which is a user of the automatic evidential document sorting apparatus 10, sorts a voucher, for example, a receipt C, of its client B.

First, proceed to step (hereinafter referred to as "ST") 1 in Fig. 4. In ST1, a person in charge at accounting firm A scans receipt C, which is the subject of processing, with a scanning device as is, and imports and obtains voucher image information, for example, "receipt image information" of receipt C.
Next, the "Receipt image information acquisition unit (program) 21" in Figure 2 operates and stores the basic information for accounting contained in the receipt image information of receipt C, for example, wording information, as text information together with the information "Receipt identification information: Receipt C" in the "Receipt image information storage unit 22" in Figure 2.

As shown in FIG. 2, the receipt image information for this receipt C includes, for example, "Today's Health (magazine title)," "Monthly Economy (magazine title)," "Note (generic product name)," and "Noda Books Co., Ltd. (publisher)." These are converted into text and stored in the "receipt image information storage unit 22" as text information together with the cleansed image information.

Next, proceed to ST2. In ST2, the content information determination unit, for example, the "first content information determination unit (program) 23" in FIG. 2, operates and references the "first registered receipt image information storage unit 24 (first analogy database)" in FIG. 2, which stores the already registered "receipt image information" and already registered evidence image information, for example, the "first already registered receipt image information."

FIG. 6 is a schematic diagram showing the state of the "first registered receipt image information storage unit (part 1)" at step ST2.
As shown in FIG. 6, for example, receipt X, receipt Y, receipt Z, etc. are stored as the first registered receipt image information.

These receipts X and the like include not only the wording information "Issued by: Hokuetsu Gas Co., Ltd." and "Contents: basic gas fee, gas usage amount," but also unique information such as "Date: 6/26" and "Receipt amount: 30,000 yen."
Furthermore, the "first registered receipt image information" of receipts X and the like is stored in association with, for example, "posted data," which is registered journalization information that has already been posted to a category or the like based on this information.

Specifically, the following applies:
(1) The "image of receipt" of receipt X and the text data of the words displayed in the image, "Hokuetsu Gas Co., Ltd. (issuer)", "basic gas charge (content of the charge)" and "gas usage charge (content of the charge)". The date information is 6/26, and the receipt amount information is 30,000 yen.
(2) "Image of receipt" of receipt Y and text data of the words displayed in the image: "Noda Books Co., Ltd. (Publisher)", "Today's Health 1 (Magazine Title)" and "Monthly Economy 1 (Magazine Title)". The date information is 7/16, and the receipt amount information is 1,500 yen.
(3) The "image of receipt" of receipt Z and the text data of the words displayed in the image, "Binder 50 (general product name),""Notebook 100 (general product name)," and "Wrapping film 50 (general product name)." The date information is 8/12, and the receipt amount information is 80,500 yen.

In addition, each of the data for "Receipt X,""ReceiptY," and "Receipt Z" is associated with the "journalized data" after the journalization process and stored in the "first registered receipt image information storage unit (first analogy database)" as follows.
(1-1) The “Accounted data” associated with “Receipt X” is “Date: 6/27 Account: Utility expenses/Cash Amount: 30,000 yen Summary: None.”
(2-1) The “Accounted-for data” associated with “Receipt Y” is “Date: 7/27 Account: Books/Cash Amount: 1,500 yen Summary: Monthly magazine subscription.”
(3-1) The "Accounted data" associated with "Receipt Z" is "Date: 8/12 Account: Purchase/Deposit Amount: 80,500 yen Summary: 50 sets of notes."

Next, proceed to ST3. In ST3, the determination unit 23 operates to compare the "text information of the receipt image information" of receipt C stored in the receipt image information storage unit 22 with the text data of receipt X, receipt Y, and receipt Z of the "registered receipt image information" in the "first registered receipt image information storage unit 24."
Specifically, the wording data of receipt C, namely "Today's Health,""MonthlyEconomy,""Note," and "Noda Books Co., Ltd.", are compared with the wording data of receipts X, Y, and Z, and it is determined whether the wording data of receipt C matches the wording data of receipts X, Y, and Z.

Next, the process proceeds to ST4. In ST4, the determination unit 23 operates to generate comparison information for each of the receipts X, Y, and Z, such as information on the degree of match, for example, information on the number of matching wording data, and store the information in the "first comparison information storage unit."
For example, the matching wording data for receipt X is “0”, for receipt Y it is “3” ("Noda Books Co., Ltd.", "Today's Health", "Monthly Economy"), and for receipt Z it is “1” ("Note"), which are stored in the “first comparison information storage unit 25” in FIG. 2.

Next, the process proceeds to ST5. In ST5, the "first candidate receipt selection unit (program) 26" in Fig. 2 operates to select a candidate receipt to be used in creating the post-transaction data for receipt C, and stores the selected candidate receipt in the "first candidate receipt storage unit 27" in Fig. 2.
Specifically, for example, "Receipt Y" which has the most text data that matches the text information of the already registered receipt image information (for example, "3"), is stored as a "candidate receipt" in the "first candidate receipt memory unit 27."

Next, proceed to ST6. In ST6, the "threshold information determination unit (program) 28" in FIG. 2 operates to determine whether the number of matches in the wording data of the candidate receipt (receipt Y) satisfies predetermined threshold information, for example, a threshold condition, and if the threshold condition is satisfied, it is stored as the "first journalization receipt" in the "first journalization receipt storage unit 29" in FIG. 2.

For example, if the number of matches is more than half of the total number of words, then in the above example, receipt C has four pieces of word data and three matches, so it meets the threshold condition and receipt Y is stored in the first journalization receipt memory unit 29 as the first journalization receipt.
In addition, the text information of this receipt Y becomes "registered basic information for journalization to be used as information to be used for journalization."

In this way, in this embodiment, a candidate receipt is designated as a "first journalization receipt" only if the number of matches in the wording data of the candidate receipt meets a predetermined threshold condition, allowing for highly accurate information.

Next, proceed to ST7. In ST7, the "first journalized data generation unit (program) 30" in FIG. 2 operates, and refers to the "first registered receipt image information storage unit 24" in FIG. 2 to change unique information such as the date and amount of the "journalized data" associated with "Receipt Y", which is the "first journalized receipt", to generate "journalized data" for "Receipt C", and store this in the "first registered receipt image information storage unit 24" in association with "Receipt C".

FIG. 7 is a schematic diagram showing the state of the "first registered receipt image information storage unit (part 2)" at step ST7.
As shown in Figure 7, the "first registered receipt image information storage unit (part 2)" differs from the "first registered receipt image information storage unit (part 1)" in Figure 6 in that the first registered receipt image information of receipt C and its sorted data are newly added and stored.

The transaction data for this receipt C is as follows: "Subject: Books/Cash", "Publisher: Noda Books Co., Ltd.", "Date: 7/16", "Receipt amount: 1,500 yen", "Summary: Monthly magazine subscription", etc.

In addition, the "unique information" here is the amount, date, and issuer, but when accounting firm A processes the journal entries of receipts for the same client, if the text data meets a certain threshold condition, there is a high possibility that the issuer is also the same, so the unique information can be only the amount and date.

As described above, according to this embodiment, the wording information such as "Today's Health" contained in the receipt image information of receipt C is determined by comparing it with the first already registered receipt image information such as the wording information of receipt X, etc., so there is no need to prepare separate information for comparison of receipt C to be bookended; it is sufficient to compare it with the first already registered voucher image information of receipt X, etc., which has already been registered as the original.
This allows easy and quick comparisons to be made.

Then, new transacted data is generated by changing only the unique information such as the date and amount, which is the text information of the transacted data associated with the corresponding first registered voucher image information, i.e., receipt Y, so that erroneous input by users, etc. can be prevented when generating new transacted data.
In addition, when selecting the wording information of the first registered receipt image information to be used as usage information for journalizing, the first registered receipt image information (receipt Y) whose number of matches with the wording information of receipt C meets a predetermined threshold condition is used as the ``receipt for journalizing.''
Therefore, it is possible to generate more accurate journalized data.

Second Embodiment
The second various information storage unit 40 shown in FIG. 3 is a schematic block diagram showing the main configuration of the "voucher automatic journalizing device 10" according to the second embodiment of the present invention.
The main configuration of this embodiment is similar to that of the first embodiment described above, so a description of the similar configuration will be omitted and the following description will focus on the differences.

As shown in Figure 3, the wording information (issuer: Noda Books Co., Ltd., etc.) of the "second registered receipt image information" in the "second registered receipt image information storage unit 41", which is basic information for registered accounting entries, is associated with "weighting information", for example, points, which differ depending on the type of the wording information.
For example, the wording information "Noda Books Co., Ltd." is rated 5 points, "Yamada Bookstore Co., Ltd." is rated 2 points, and "Today's Health,""MonthlyEconomy," and "Note" are rated 1 point each.

The number of points is preset depending on the importance of the wording information in the accounting firm or the like that processes the document.
The number of points may be configured to be automatically changed depending on the processing frequency.
For example, when the number of times that the same text information matches reaches a predetermined number, the number of points may be increased.
In this case, among the word information that functions as a keyword, the more frequent and important the word information is, the higher the points are given, so that journalizing data with higher accuracy can be created.

In the present embodiment, an example in which points are assigned to the wording information in advance will be described below.
Specifically, as shown in FIG. 3, a point number is assigned to each piece of wording information, and the information is stored in the second registered receipt image information storage unit (second analogy database) 41 .

FIG. 5 is a schematic flow chart showing the main operations of the automatic document sorting device 10 according to the present embodiment.
The details will be explained below according to the flowchart.

In this embodiment, first, in ST11, a process similar to ST1 in the above-described first embodiment is executed.
That is, "the receipt C to be processed is scanned, "receipt image information" is acquired, and the receipt image information acquisition unit 21 in FIG. 2 operates to store the text information contained in this receipt image information, for example, "Publisher: Noda Books Co., Ltd.", "Contents: Today's Health, Monthly Economy, Notes, etc." as text information together with "receipt identification information: receipt C" in the "receipt image information storage unit 22" in FIG. 2.

Next, proceed to ST12. In ST12, a content information determination unit, for example, the "second content information determination unit (program) 42" in Fig. 3 operates to refer to the "second already registered receipt image information storage unit (second analogy database) 41" which stores the already registered "receipt image information" and the already registered evidence image information, for example, the "second already registered receipt image information."
As described above, this second registered receipt image information stores the "number of points" in association with the wording information such as the issuer and contents.

Then, proceed to ST13. In ST13, the judgment unit 42 operates to compare the "text information of receipt image information" of receipt C stored in the "receipt image information storage unit 22" in FIG. 2 with the text data of receipt V and receipt W in the "second registered receipt image information" of the "second registered receipt image information storage unit 41" in FIG. 3.

Specifically, the text data of receipt C, "Today's Health," "Monthly Economy," "Note," and "Noda Books Co., Ltd." are compared with the text data of receipts V and W to determine whether the text data of receipt C matches the text data of receipts V and W.

Next, proceed to ST14. In ST14, the judgment unit 42 operates to generate comparison information for each of receipts V and W, such as information on the degree of match, for example, information on the number of matching wording data, and store this in the "second comparison information storage unit 43" in Figure 3, and also store the total "number of points" for each receipt.

For example, the matching wording data for receipt V is "2" (Noda Books Co., Ltd.: 5 points, Today's Health: 1 point, total points 6 points), and for receipt W it is "3" (Today's Health: 1 point, Monthly Economy: 1 point, Note: 1 point, total points 3 points), which are stored in the "second comparison information storage unit 43".

Next, proceed to ST15. In ST15, the "second candidate receipt selection unit (program) 44" in FIG. 3 operates to select a candidate receipt to be used to create the post-account data for receipt C, and stores it in the "second candidate receipt storage unit 45" in FIG. 3.

Specifically, "Receipt V" is the one with the highest "total points" of "6" in the wording data that matches the wording information of the already registered receipt image information, and is stored as a "candidate receipt" in the "second candidate receipt storage unit 45."

Next, proceed to ST16. In ST16, the points determination unit (program) 46 in FIG. 3 operates to determine whether the "total points" in the text data of the candidate receipt (receipt V) meets a predetermined threshold condition, and if it does, stores it as a "second journalization receipt" in the "second journalization receipt storage unit 47" in FIG. 3.

For example, if the "threshold" for the "total points" is set to a point number of 5 or more for the entire text information, in the above example, receipt V has a "total points" of "6", which satisfies the threshold condition, and receipt V is stored as a "second journalizing receipt" in the "second journalizing receipt memory unit 47" in Figure 3.
The text information of this receipt V is an example of "registered basic information for journalization to be used as information to be used for journalization."

Next, proceed to ST17. In ST17, the "second journalized data generation unit (program) 48" in FIG. 3 operates to change unique information such as the date and amount of the "journalized data" associated with "Receipt V", which is the "second journalized receipt", to generate "journalized data" for "Receipt C", and store this in the "second registered receipt image information storage unit 41" in FIG. 3 in association with "Receipt C".

As described above, according to the second embodiment, the weighting information such as the point number, which is the weighting information for the wording information of the issuer of the receipt, the title of the magazine, the general name of the product, etc., is different, so that when generating comparison information, it is possible to make a comparison that emphasizes the wording information with a high point number.
This makes it possible to generate more accurate journalized data.

In addition, when the number of points for the text information is automatically changed depending on the processing frequency of the text information of the receipt image information, the system is configured to "learn" so as to automatically generate "comparison information" that is in line with the accounting processing trends of users, such as accounting firms, that use the automatic accounting device.
Therefore, the automatic voucher sorting device 10 is easy for users to use.

Furthermore, when the automatic document sorting device 10 having the configuration of the first or second embodiment described above is used in an accounting firm or the like, the accounting firm's past trends in sorting processing, etc. are learned, and the accuracy of the sorting information of the receipt image information improves with repeated processing.

(Modifications of the first and second embodiments)
In the first and second embodiments described above, if the basic information for journalization stored in the "Receipt image information storage unit 22" in Figure 2, for example, the wording information for "issuer" on receipt C, is unclear due to being unclear or the like, the data for "issuer" in the "journalized data" written in the "First registered receipt image information storage unit (part 2)" in Figure 7 will be "unknown," and the journalized data will be incomplete.

Therefore, in this embodiment, even if the "issuer" on receipt C is unclear, the issuer of receipt C can be estimated from the already stored color of the receipt, the arrangement of items, the shape and size of the ruled rectangle, etc.
Specifically, the device 10 has a "form information storage unit" that stores information such as the color and item arrangement of receipts obtained in the past, and the shape and size of ruled rectangles.
FIG. 8 is a schematic diagram showing an example of morphological information stored in the morphological information storage unit.
In the example shown in FIG. 8, information on the arrangement of items on a receipt is stored.
In the present invention, information is not limited to the layout of items on the receipt, the shape and size of the ruled rectangle, etc., but may also include information on the color of each part, etc.

Therefore, in this embodiment, when the "issuer estimation unit (program)" of the device 10 operates and determines that the "issuer" information of receipt C is unclear, it refers to the above-mentioned "form information storage unit" to extract the "receipt item arrangement state" that is the same as or similar to the "receipt item arrangement state" of receipt C, and estimates the "issuer" information of receipt C based on the "issuer" information of the receipt linked to that "receipt item arrangement state."

Therefore, even if basic information for accounting (e.g., information on the issuer, etc.) contained in evidence image information such as receipt C cannot be obtained, information on the issuer, etc. can be identified from shape information, etc., and new accounting data can be automatically generated, thereby preventing the occurrence of incomplete accounting data.

In the above embodiment, the present invention has been described with reference to an example in which the invention is realized as a device, but the invention is not limited to this, and the invention may be distributed as a program that can be executed by a computer and stored on a storage medium such as a magnetic disk, optical disk (CD-ROM, DVD, etc.), magneto-optical disk (MO), or semiconductor memory.

The storage medium may be any medium that can store a program and is computer-readable. There are no particular limitations on the storage format of the storage medium.

In addition, an OS (operating system), database management software, network software, or other MW (middleware) running on a computer may execute some of the processes for implementing this embodiment based on instructions from a program installed on the computer from a storage medium.

Furthermore, the storage medium in the present invention is not limited to a medium independent of a computer, but also includes a storage medium that stores or temporarily stores a program downloaded from a LAN, the Internet, etc.

The computer of the present invention only needs to execute each process in this embodiment based on a program stored in a storage medium, and may be a device consisting of a single personal computer, or a system in which multiple devices are connected to a network.

In addition, the computer in this invention is not limited to a personal computer, but also includes an arithmetic processing unit, a microcomputer, etc., included in information processing equipment, and is a general term for equipment or devices that can realize the functions of this invention by a program.

The above describes an embodiment of the present invention. However, the present invention is not limited to the above embodiment, and various modifications can be made without departing from the scope of the claims. The configurations of the above embodiment can be partially omitted or arbitrarily combined in a different way than described above.

10: Automatic document sorting device, 11: Control unit, 12: Communication device, 13: Display, 14: Various information input device, 20: First various information storage unit, 21: Receipt image information acquisition unit, 22: Receipt image information storage unit, 23: First content information judgment unit, 24: First registered receipt image information storage unit, 25: First comparison information storage unit, 26: First candidate receipt selection unit, 27: First candidate receipt storage unit, 28:・Threshold information judgment unit, 29...First journalization receipt storage unit, 30...First journalized data generation unit, 40...Second various information storage unit, 41...Second registered receipt image information storage unit, 42...Second content information judgment unit, 43...Second comparison information storage unit, 44...Second candidate receipt selection unit, 45...Second candidate receipt storage unit, 46...Point judgment unit, 47...Second journalization receipt storage unit, 48...Second journalized data generation unit

Claims (6)

The image information includes evidence image information, which is image information related to evidence, and registered evidence image information, which is the image information of the evidence that has already been registered;
The content information determination unit generates comparison information by comparing the journalization basic information included in the evidence image information with the registered journalization basic information included in the registered evidence image information;
When the content information determination unit determines that the comparison information matches the threshold information, the content information determination unit selects the registered basic information for journalization to be used as usage information for journalization,
Extracting registered journal information associated with the registered evidence image information of the selected registered journalization basic information, acquiring unique information of the evidence image information, and rewriting only the unique information of the registered journalization information to generate journalization information of the evidence image information ;
a shape information storage unit for storing information on the color and item arrangement of the voucher obtained in the past, and the shape and size of the ruled rectangle;
The unique information is information on the amount, date, and issuer,
This automatic sorting device is characterized in that even if it is not possible to obtain information on the issuer, the device is configured to estimate information on the issuer of the voucher from the color of the voucher, the arrangement of items, and the shape and size of the ruled rectangle stored in the shape information memory unit. 2. The automatic journalizing apparatus according to claim 1 , wherein the pre-registered basic information for journalizing is associated with different weighting information depending on the type of the information . 3. The automatic journalizing device according to claim 2 , wherein the weighting information associated with the pre-registered journalizing basic information is automatically changed depending on a processing frequency of the pre-registered journalizing basic information . 3. The automatic journalizing device according to claim 1 , wherein the journalizing information in the evidence image information is journalizing information in an accounting firm . A method for controlling an automatic sorting apparatus having evidence image information, which is image information related to evidence, and pre-registered evidence image information, which is pre-registered evidence image information, comprising:
The content information determination unit generates comparison information by comparing the journalization basic information included in the evidence image information with the registered journalization basic information included in the registered evidence image information;
Select the registered basic information for journal entry to be used as information for journal entry based on the comparison information,
Extract the registered journal information associated with the registered evidence image information of the selected registered journalization basic information, and obtain unique information, which is information on the amount, date, and issuer of the evidence image information, and rewrite only the unique information of the registered journal information to generate journal information for the evidence image information;
The color of the voucher, the arrangement of items, and the shape and size of the ruled rectangle are stored in a shape information storage unit.
A control method for an automatic sorting device, characterized in that even if it is not possible to obtain information on the issuer, information on the issuer of the voucher is estimated from the color of the voucher, the arrangement of items, and the shape and size of the ruled rectangle stored in the shape information memory unit . An automatic sorting device having evidence image information, which is image information related to evidence, and pre-registered evidence image information, which is pre-registered evidence image information,
A function of a content information determination unit comparing the journalization basic information included in the evidence image information with the registered journalization basic information included in the registered evidence image information to generate comparison information ;
A function of selecting the already registered basic information for journal entry to be used as information for journal entry based on the comparison information;
A function of extracting the already registered journal information associated with the already registered evidence image information of the selected already registered journalization basic information, acquiring unique information that is information on the amount, date, and issuer of the evidence image information, and rewriting only the unique information of the already registered journal information to generate journal information of the evidence image information;
A function of storing information on the color and item arrangement of the voucher obtained in the past, and the shape and size of the ruled rectangle in a shape information storage unit;
A control program for an automatic sorting device for realizing a function of estimating information on the issuer of the voucher from the color, item arrangement, and shape and size of the ruled rectangle of the voucher stored in the shape information memory unit, even if the information on the issuer cannot be obtained.

Images