Accounting information processing system, accounting information processing method, and accounting information processing program

The system automates the sorting of deposit passbook images using a storage unit, determination unit, and machine learning to identify and group images, reducing labor for tax accountants.

JP2026061573AActive Publication Date: 2026-04-09FAST ACCOUNTING INC
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2026-04-09

AI Technical Summary

Technical Problem

Sorting large quantities of deposit passbook images from multiple customers is labor-intensive for tax accountants.

Method used

An accounting information processing system that includes a bank passbook image storage unit, a determination unit to identify matching passbook images, and a storage unit to associate images and determination results, utilizing transaction information, font, and machine learning for automatic identification.

Benefits of technology

Automatically determines which deposit passbook images belong to the same passbook, reducing the effort required for sorting and organizing images.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026061573000001_ABST
    Figure 2026061573000001_ABST
Patent Text Reader

Abstract

This reduces the effort required for tax officials to organize images of bank passbooks. [Solution] This system includes a bank passbook image storage unit 1 that stores multiple bank passbook images, a bank passbook image identity determination unit 2 that determines whether the multiple bank passbook images stored in the bank passbook image storage unit 1 are images of the same bank passbook and outputs the determination result, and a bank passbook image identity storage unit 3 that stores the bank passbook images and the determination result output by the bank passbook image identity determination unit in association.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an accounting information processing system, an accounting information processing method, and an accounting information processing program.

Background Art

[0002] In some cases, an image of a deposit passbook of a financial institution such as a bank is transmitted from a customer to a tax accountant's office or an accounting office via email or communication software. At a tax accountant's office or the like, a tax accountant looks at the image of the deposit passbook transmitted from the customer and performs operations such as inputting information necessary for taxation into an accounting information processing system.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When images of deposit passbooks are transmitted in large quantities from a large number of customers, there is a problem that it takes effort to sort out which deposit passbook the image is of.

[0005] An object of the present invention is to provide an accounting information processing system, an accounting information processing method, and an accounting information processing program that can automatically determine which deposit passbook an account book image is, and reduce the labor required for sorting out images of deposit passbooks by tax accountants.

Means for Solving the Problems

[0006] To solve the above problems, an accounting information processing system according to an embodiment of the present invention includes: a bank passbook image storage unit that stores a plurality of bank passbook images; a bank passbook identity determination unit that determines whether the plurality of bank passbook images stored in the bank passbook image storage unit are bank passbook images relating to the same bank passbook and outputs a determination result; and a bank passbook identity storage unit that stores the bank passbook images and the determination result output by the bank passbook identity determination unit in association.

[0007] Furthermore, it is preferable that the system includes a transaction information output unit that outputs information regarding the date and balance of each transaction recorded in the plurality of passbook images, and that the passbook identity determination unit determines whether the plurality of passbook images are passbook images relating to the same passbook based on the transaction date and balance output by the transaction information output unit, and outputs the determination result.

[0008] Furthermore, it is preferable to include a sequence output unit that outputs the order of multiple deposit passbook images based on the transaction date output by the transaction information output unit, for multiple deposit passbook images that the passbook identity determination unit has determined to be deposit passbook images relating to the same deposit passbook.

[0009] Furthermore, it is preferable that the deposit passbook identity determination unit determines whether the multiple deposit passbook images are deposit passbook images relating to the same deposit passbook based on the color, item name or / and the format of the characters recorded in the deposit passbook images, and outputs the determination result.

[0010] Furthermore, it has a machine learning model that, when given multiple bank passbook images as input, outputs whether or not the multiple bank passbook images pertain to the same bank passbook. Preferably, the deposit passbook identity determination unit inputs a plurality of deposit passbook images stored in the deposit passbook image storage unit to the learning model, determines whether the plurality of deposit passbook images are deposit passbook images relating to the same deposit passbook, and outputs the determination result.

[0011] Furthermore, if the deposit passbook image contains handwritten entries, it is preferable to have a memo recognition unit that applies character recognition technology to the handwritten entries to recognize the characters and stores them in association with transactions near the handwritten entries.

[0012] In another embodiment of the present invention, the accounting information processing program causes a computer to execute: a deposit passbook image storage step of storing a plurality of deposit passbook images; a deposit passbook identity determination step of determining whether the plurality of deposit passbook images stored in the deposit passbook image storage step are deposit passbook images relating to the same deposit passbook and outputting the determination result; and a deposit passbook identity storage step of storing the deposit passbook images and the determination result output by the deposit passbook identity determination step in association.

[0013] In another embodiment of the present invention, the accounting information processing method comprises: a deposit passbook image storage step performed by a computer to store a plurality of deposit passbook images; a deposit passbook identity determination step that determines whether the plurality of deposit passbook images stored in the deposit passbook image storage step are deposit passbook images relating to the same deposit passbook and outputs the determination result; and a deposit passbook identity storage step that stores the deposit passbook images and the determination result output by the deposit passbook identity determination unit in association. [Effects of the Invention]

[0014] According to the present invention, it is possible to automatically determine which bank passbook an image belongs to, thereby reducing the effort required for tax officials to organize images of bank passbooks. [Brief explanation of the drawing]

[0015] [Figure 1] This figure shows the configuration of an accounting information processing system according to an embodiment of the present invention. [Figure 2] This diagram shows the configuration of a computer used in implementing the present invention. [Figure 3] It is a diagram showing the information processing of Example 1. [Figure 4A] It is a diagram showing the deposit passbook image A. [Figure 4B] It is a diagram showing the deposit passbook image B. [Figure 4C] It is a diagram showing the deposit passbook image C. [Figure 5] It is a diagram showing the deposit passbook image D. [Figure 6] It is a diagram showing a part of the data obtained by reading the deposit passbook image D by OCR.

Mode for Carrying Out the Invention

[0016] FIG. 1 is a diagram showing the configuration of an accounting information processing system according to an embodiment of the present invention. This system includes a deposit passbook image storage unit 1 that stores a plurality of deposit passbook images, a deposit passbook identity determination unit 2 that determines whether the plurality of deposit passbook images stored in the deposit passbook image storage unit 1 are images of the same deposit passbook based on the plurality of deposit passbook images and outputs a determination result, and a deposit passbook identity storage unit 3 that stores the deposit passbook images and the determination result output by the deposit passbook identity determination unit in association with each other.

[0017] In this specification, a "deposit passbook" refers to a passbook created by a financial institution or the like for the purpose of continuously recording and proving the receipt and payment of continuous deposits between the depositor or saver, and includes a savings passbook. FIGS. 4A to 4C are examples of images of deposit passbooks. The items, font, font size, layout, background color, etc. of the deposit passbook vary depending on the financial institution that issues the deposit passbook.

[0018] A "deposit passbook image" is an image of a deposit passbook obtained by photographing a deposit passbook with a digital camera or scanning it with a scanner.

[0019] In this specification, the deposit passbook image identity determination unit not only determines whether a plurality of deposit passbook images are images related to the same deposit passbook, but may also determine that there is a high possibility that they are images related to the same deposit passbook. It is also possible to display on the display that there is a high possibility that a certain deposit passbook image is related to the same deposit passbook as another deposit passbook image, and based on the input of the user's decision, determine that it is an image related to the same deposit passbook.

[0020] The information processing by this system is executed by a computer. Figure 2 is a diagram showing the configuration of computer 20 used in this system. The storage unit 21 stores deposit passbook images and functions as a deposit passbook storage unit. The arithmetic unit 22 applies character recognition technology (OCR, Optical Character Recognition / Reader) to the deposit passbook image and reads transaction information regarding the date of each transaction, payment amount, deposit amount, balance, etc. recorded in the deposit passbook image. Further, the arithmetic unit 22 functions as a deposit passbook image identity determination unit that determines whether a plurality of deposit passbook images are of the same deposit passbook based on the read transaction information. The output unit 24 is a device that outputs the information stored in the storage unit 21, such as a display, speaker, etc. The input unit 23 is a device for the user to input information into the computer, such as a keyboard, mouse, camera, microphone, etc. The communication unit 25 is a device for communicating information with devices outside the computer.

Example

[0021] Hereinafter, Example 1 of the present invention will be described using Figure 3. First, the deposit passbook image transmitted from the customer is input into the accounting information processing system and stored in the deposit passbook image storage unit. Deposit passbook images A to C in Figures 4A to 4C are examples of deposit passbook images.

[0022] Next, the bank passbook image is read using AI-OCR, and transaction information such as the transaction date, description, payment amount, deposit amount, and net balance for each transaction recorded in the bank passbook image is recognized and output (S11). Specifically, the bank passbook image is input to a learning model that has been trained with training data to output information corresponding to items such as "transaction date," "description," "payment amount," "deposit amount," and "net balance," and the information corresponding to these items is output. For example, when AI-OCR is applied to bank passbook image A (Figure 4A), information is output that 104,000 yen was withdrawn from an ATM on January 4, 2024, and the net account balance became 9,000,000 yen. Information is output that bank passbook image A contains transactions recorded from January 4, 2024 to February 22, 2024. Furthermore, when AI-OCR is applied to passbook image B (Figure 4B), the output indicates that passbook image B contains information indicating that transactions from February 20, 2024 to March 31, 2024 are recorded.

[0023] Then, the transaction information of bank passbook image A and bank passbook image B is compared, and it is automatically determined whether bank passbook image A and bank passbook image B are images of the same bank passbook. Specifically, the transaction period recorded in bank passbook image A and the transaction period recorded in bank passbook image B overlap from February 20, 2024 to February 22, 2024, and the balance of bank passbook image A on February 20, 2024 is 9,616,000 yen, while the balance of bank passbook image B on February 20, 2024 is 1,026,259 yen, and the balances on the same day do not match, so it is determined that bank passbook image A and bank passbook image B are bank passbook images of different bank passes.

[0024] When AI-OCR is applied to bank passbook image C (Figure 4C), the output indicates that bank passbook image C records transactions from February 25, 2024 to March 22, 2024. There is no overlap between the transaction periods of bank passbook image A and bank passbook image C. The last transaction date recorded in bank passbook image A is February 22, 2024, which is close to the first transaction date recorded in bank passbook image C, February 25, 2024. Furthermore, the balance in bank passbook image A on February 22, 2024 is 9,571,000 yen, and in the transaction on February 25, 2024, in bank passbook image B, there was a payment of 1,000,000 yen, leaving a balance of 8,571,000 yen. Considering the payment amount and balance, it is presumed that bank passbook image A and bank passbook image B are images relating to the same bank passbook and are images of consecutive pages from the same bank passbook. Therefore, it is determined that bank passbook A and bank passbook B are images relating to the same bank passbook (S12).

[0025] Based on these determination results, multiple bank passbook images relating to the same bank passbook are grouped and stored in the memory unit (S13). For example, bank passbook image A and bank passbook image C are labeled with bank passbook number: "Bank Passbook 001", etc., and stored. Furthermore, labels such as "Bank Passbook 001" being named "A Corporation" and relating to "A Bank" may also be added. Alternatively, the order of the bank passbook images may be determined based on the transaction date, etc., and each bank passbook image may be assigned an identification number that indicates the page number, such as "Bank Passbook 001-001" for bank passbook image A and "Bank Passbook 001-002" for bank passbook image C, and displayed on the display in chronological order for each identical bank passbook. [Examples]

[0026] The following describes the processing procedure for Example 2 of the present invention. Explanations common to Example 1 will be omitted. The font used in bank passbook image A (Figure 4A) is "MS Mincho," and the only item names are "Date," "Description," "Payment Amount," "Deposit Amount," and "Balance." The backgrounds for "Description" and "Balance" are pink. In this specification, the font and size of the characters are referred to as the "formatting" of the characters.

[0027] On the other hand, the font in bank passbook image B (Figure 4B) is "MS Gothic," the item names contain "symbols" that are not present in bank passbook image A, and the background of "Balance" is light green. Therefore, bank passbook image A and bank passbook image B differ in font, items, background color, etc., indicating that they were issued by different financial institutions. Thus, bank passbook image A and bank passbook image B are determined to be images of different bank passbooks. If information such as the font, items, and background color of each financial institution is recorded in advance, the name of the financial institution that issued the bank passbook can be identified from the bank passbook image. Alternatively, the name of the financial institution that issued the bank passbook can be identified based on information regarding the dimensions such as the width and height of each item, and the layout such as the position.

[0028] Both bank passbook image A and bank passbook image B share the same font ("MS Mincho"), item names ("Date", "Description", "Payment Amount", "Amount Received", "Balance"), and a pink background for "Description" and "Balance". Therefore, it can be inferred that bank passbook image A and bank passbook image B are images of bank passbooks issued by the same financial institution. While images of bank passbooks issued by the same financial institution do not necessarily mean they are images of the same bank passbook, this information is useful for determining the identity of bank passbooks. [Examples]

[0029] The following describes the processing procedure for Example 3 of the present invention. Explanations common to Examples 1 and 2 will be omitted.

[0030] In this embodiment, a machine learning model is constructed that, upon input of multiple bank passbook images, for example, two bank passbook images, outputs whether or not the two input bank passbook images pertain to the same bank passbook.

[0031] Next, bank passbook image A and bank passbook image C are input into the constructed learning model. Since bank passbook image A and bank passbook image C share the characteristic of having phone bills deducted on the 15th of each month, electricity bills deducted on the 20th of each month, water bills deducted on the 22nd of each month, and salary payments on the 25th of each month, the learning model outputs and determines that bank passbook image A and bank passbook image C are images of the same bank passbook. [Examples]

[0032] The following describes the processing procedure for Embodiment 4 of the present invention. Explanations common to the embodiments described so far will be omitted.

[0033] Figure 5 is a bank passbook image D. The entry for February 28, 2024, has "K Deposit" written in handwriting. Figure 6 is a portion of the data read from bank passbook image D using AI-OCR. In the transaction for "February 28, 2024," there is a deposit of 2,500,000 yen, and the "Handwritten Memo" field has "A Deposit" entered. Therefore, it can be inferred that the 2,500,000 yen deposit is either from a business partner named A or payment for a product named A, providing the user with a clue to determine the cause of the deposit.

[0034] The embodiments and examples of the present invention described above are illustrative examples of the present invention, and the technical scope of the present invention is not limited to the above embodiments, etc. The configuration of the present invention can be implemented with various modifications within the scope of the claims. [Explanation of symbols]

[0035] 1. Bankbook image storage unit 2. Deposit passbook identity determination unit 3. Deposit passbook identity storage section 20 Computers 21 Memory section 22 Arithmetic section 23 Input section 24 Output section 25 Communications Department

Claims

1. A deposit passbook image storage unit that stores multiple deposit passbook images, A deposit passbook identity determination unit determines whether a plurality of deposit passbook images stored in the deposit passbook image storage unit are deposit passbook images relating to the same deposit passbook, and outputs the determination result. An accounting information processing system comprising: a deposit passbook identity storage unit that stores the deposit passbook image and the determination result output by the deposit passbook identity determination unit in association with each other.

2. The system includes a transaction information output unit that outputs information regarding the date and balance of each transaction recorded in the aforementioned plurality of deposit passbook images, The accounting information processing system according to claim 1, wherein the deposit passbook identity determination unit determines whether multiple deposit passbook images are deposit passbook images relating to the same deposit passbook based on the transaction date and balance output by the transaction information output unit, and outputs the determination result.

3. The accounting information processing system according to claim 2, further comprising a sequence output unit that outputs the order of a plurality of deposit passbook images based on the date of each transaction output by the transaction information output unit, for a plurality of deposit passbook images that the passbook identity determination unit has determined to be deposit passbook images relating to the same deposit passbook.

4. The account information processing system according to claim 1, wherein the deposit passbook identity determination unit determines whether the multiple deposit passbook images are deposit passbook images relating to the same deposit passbook based on the color, item name, and / or the format of the characters recorded in the deposit passbook images, and outputs the determination result.

5. The system has a machine learning model that, upon inputting multiple bank passbook images, outputs whether or not those multiple bank passbook images pertain to the same bank passbook. The accounting information processing system according to claim 1, wherein the deposit passbook identity determination unit inputs a plurality of deposit passbook images stored in the deposit passbook image storage unit to the learning model, determines whether the plurality of deposit passbook images are deposit passbook images relating to the same deposit passbook, and outputs the determination result.

6. The accounting information processing system according to claim 1, further comprising a memo recognition unit that, if the deposit passbook image contains handwritten entries, applies character recognition technology to the handwritten entries to recognize the characters and stores them in association with transactions near the handwritten entries.

7. On the computer, A deposit passbook image storage step that stores multiple deposit passbook images, A deposit passbook identity determination step that determines whether the multiple deposit passbook images stored in the deposit passbook image storage step are deposit passbook images relating to the same deposit passbook, and outputs the determination result, A deposit passbook identity storage step that stores the deposit passbook image and the determination result output by the deposit passbook identity determination step in association with each other, An accounting information processing program that executes this process.

8. A computer executes A deposit passbook image storage step that stores multiple deposit passbook images, A deposit passbook identity determination step that determines whether the multiple deposit passbook images stored in the deposit passbook image storage step are deposit passbook images relating to the same deposit passbook, and outputs the determination result, A deposit passbook identity storage step involves associating and storing the deposit passbook image and the determination result output by the deposit passbook identity determination unit, An accounting information processing method comprising the following features.