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

Abstract

This can reduce the burden on the person in charge at the receiving end of the evidence to verify it. [Solution] The system includes a document image upload unit 11 for uploading document images of the document, a text information output unit 12 for applying character recognition technology to the document images uploaded to the document image upload unit 11 and outputting text information recorded in the document images, a display control unit 13 for displaying the text information output by the text information output unit 12 on a display unit used by the issuer of the document, and a text information correction reception unit 14 for receiving corrections to the text information displayed by the display control unit 13.

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] There is a technology in which AI-OCR (Optical Character Recognition) processing is performed on the electronic data of vouchers such as invoices on the recipient side of the vouchers and converted into text information, thereby reducing the labor for a person to input information related to the vouchers into the 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] Even when using AI-OCR, it is not possible to perfectly read the information on the vouchers, and there is a possibility of misreading. Therefore, the person in charge on the recipient side of the vouchers must confirm whether the reading result of AI-OCR is correct, and if there is misread information, it must be corrected, which imposes a burden on the person in charge on the recipient side. Also, if there is incorrect information in the reading result of AI-OCR and only the issuer of the vouchers knows what the correct information is, it may be necessary to inquire the issuer of the vouchers. In such a case, there is a burden of inquiring the issuer of the vouchers, and there is a problem that the efficiency of the accounting process on the recipient side decreases. 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 reduce the burden of confirmation for the recipient of the vouchers. [Means for solving the problem]

[0005] To solve the above problems, an accounting information processing system according to an embodiment of the present invention comprises: a document image upload unit for uploading document images of documents; a text information output unit for applying character recognition technology to the document images uploaded to the document image upload unit and outputting text information recorded in the document images; a display control unit for displaying the text information output by the text information output unit on the display unit of a document issuer terminal used by the issuer of the document; and a text information correction reception unit for receiving corrections to the text information displayed by the display unit.

[0006] Furthermore, the system may include a text information normality determination unit that outputs a determination result of whether the text information is normal or not based on the text information output by the text information output unit, and the display control unit may display a message to the effect that the text information normality determination unit has determined is not normal on the display unit of the certificate issuer terminal used by the certificate issuer.

[0007] Furthermore, the system may include a text information normality determination unit that outputs a determination result of whether the text information is normal or not based on the text information output by the text information output unit. If the text information normality determination unit determines that the text information is not normal, the display control unit will display the text information output by the text information output unit on the display unit of the terminal used by the issuer of the document. If the text information normality determination unit determines that the text information is normal, the display control unit will transmit the document image to the recipient of the document image.

[0008] Furthermore, the text information normality determination unit may output a determination result of whether or not the text information is normal, based on the text information relating to the amount output by the text information output unit.

[0009] Furthermore, if the display control unit does not have the text information regarding the name of the document issuer output by the text information output unit recorded in the recipient's customer master, it may also have the display control unit display a message to that effect on the display unit of the document issuer's terminal used by the document issuer.

[0010] An accounting information processing method according to an embodiment of the present invention comprises: a document image upload step, which is performed by a computer to upload a document image of a document; a text information output step, which applies character recognition technology to the document image uploaded in the document image upload step to output text information recorded in the document image; a display control step, which displays the text information output in the text information output step on the display unit of a document issuer terminal used by the issuer of the document; and a text information correction acceptance step, which accepts corrections to the text information displayed in the display control step.

[0011] An accounting information processing program according to an embodiment of the present invention causes a computer to execute: a document image upload step of uploading a document image of a document; a text information output step of applying character recognition technology to the document image uploaded in the document image upload step and outputting text information recorded in the document image; a display control step of displaying the text information output in the text information output step on the display unit of a document issuer terminal used by the issuer of the document; and a text information correction acceptance step of accepting corrections to the text information displayed in the display control step. [Effects of the Invention]

[0012] This improves the accuracy of the document data received by the recipient of the document, reducing the burden on the person in charge at the recipient's end to verify the documents. Furthermore, it reduces the effort required for the person in charge at the recipient's end to contact the issuer of the document, thereby improving the efficiency of accounting processing at the recipient's end. [Brief explanation of the drawing]

[0013] [Figure 1] This figure shows the configuration of an accounting information processing system according to an embodiment of the present invention. [Figure 2] This figure shows the configuration of a computer used in embodiments of the present invention. [Figure 3] This diagram shows the information flow in Example 1. [Figure 4] This diagram shows a flowchart of the information processing in Example 1. [Figure 5] This diagram shows the invoice image upload screen displayed on the invoice issuer's system. [Figure 6] This diagram shows the display screen of the invoice issuer's system. [Modes for carrying out the invention]

[0014] Figure 1 shows the configuration of an accounting information processing system according to an embodiment of the present invention. The system comprises a document image upload unit 11 for uploading document images of documents, a text information output unit 12 for applying character recognition technology to the document images uploaded to the document image upload unit 11 and outputting text information recorded in the document images, a display control unit 13 for displaying the text information output by the text information output unit 12 on a display unit used by the issuer of the document, and a text information correction reception unit 14 for receiving corrections to the text information displayed by the display control unit 13.

[0015] In this specification, "proof" means a document or electronic file containing information about a transaction issued by one party to the other party. Examples of "proof" include quotations, purchase orders, delivery notes, inspection reports, and invoices.

[0016] In this specification, "text information" refers to information consisting of letters, numbers, symbols, etc.

[0017] The information processing of the present invention is executed using a computer. Figure 2 is a diagram showing the configuration of a computer 20 used in an embodiment of the present invention. The storage unit 21 is a device having a function of storing information, and examples thereof include a hard disk, an SSD, a semiconductor memory, and the like. The storage unit 21 stores voucher images, text information, programs, and the like. The arithmetic unit 22 is a device having an arithmetic function, and examples thereof include a CPU and the like. The arithmetic unit 22 functions as a text information output unit 12, a display control unit 13, and the like. The input unit 23 is a device for inputting information into the computer 20, and examples thereof include a keyboard, a mouse, a camera, a microphone, and the like. The output unit 24 is a device for outputting the information stored in the storage unit 21, and examples thereof include a display, a speaker, and the like. The communication unit 25 is a device for performing input / output of information with the outside of the computer.

Example

[0018] Hereinafter, Example 1 of the present invention will be described. In this example, an example of a claim form will be described. Figure 3 is a diagram showing the flow of information in this example. Also, Figure 4 is a flowchart of the information processing in this example.

[0019] The issuer of the claim form uses an external uploader 321 to upload and store the claim form image in the storage unit of the system 32 on the claim form recipient side (S11). The external uploader 321 is a function for uploading the claim form image from the system 31 on the claim form issuer side to the storage unit of the system 32 on the claim form recipient side. In recent years, PDF format electronic data has been the mainstream for claim form images. Figure 5 is a diagram showing the upload screen of the claim form image displayed on the display unit of the system on the claim form issuer side. When an operation is performed to drag the claim form image to be uploaded and drop it in the area displayed as "Drag the claim form image here." at 51 and press the send button 52, the claim form image is stored in the storage unit of the system 32 on the claim form recipient side. However, at this stage, the person in charge on the claim form recipient side still cannot view the uploaded claim form image.

[0020] AI-OCR (322) is applied to the claim image uploaded to the recipient-side system, and the text information recorded in the claim image is analyzed and output (S12). Specifically, when the claim image is input, the claim image is input into a learning model that has been machine-learned with teacher data (training data) so that information corresponding to predetermined items such as "date", "amount", "recipient name", "issuer" is output, and information corresponding to items such as "date", "amount", "recipient name", "issuer" of the claim image is output. Then, the text information of the claim image and the claim image are simultaneously displayed on the screen of the display unit 311 of the claim issuer-side system (S13).

[0021] Figure 6 is the display screen 60 of the claim issuer-side system. On the left side of the display screen 60, information on "date", "amount", "recipient name", and "issuer" of the claim image, which is the result of applying AI-OCR to the claim image, is displayed. These information can be modified by the user. Also, on the right side of the display screen 60, the claim image 61 is displayed. By simultaneously displaying the OCR reading result of AI and the claim image 61 on one display screen, the person in charge on the claim issuer side can easily discover misrecognition of AI-OCR.

[0022] The sum of the amounts for "2024-001 New Patent Application," "2021-013 Response to Rejection Grounds," and "2023-021 Request for Examination" in invoice image 61 is 650,000 yen, but the "Subtotal" in invoice image 61 displays "640,000," indicating that an incorrect amount has been entered on the invoice. In cases like this, where there is a discrepancy in the amount stated on the invoice, a mark 62 appears to the right of "Amount (yen)" on the left side, indicating a warning that the AI-OCR reading result may be incorrect, and to the right of that, an explanation of the reason for the warning, "Amount does not match," is displayed. This allows the person in charge at the invoice issuer to notice the error in the amount on the invoice and correct it by entering the correct value using the input unit 312 (S14). In the case of Figure 6, since there is no error in the AI-OCR reading result of the amount, and the amount stated on invoice image 61 itself is incorrect, it is also possible to cancel by pressing the "Cancel" button 67 and create the invoice again.

[0023] Other things this system could check include verifying whether the required information for a qualified invoice under the invoice system is included on the invoice. If there are any omissions, a warning mark and the missing items could be displayed on the invoice issuer's system screen.

[0024] Furthermore, it is conceivable that the system could verify whether the transactions listed on the invoice match the transactions in the order data, and if no corresponding transactions exist in the order data, display a warning mark along with a message to that effect on the invoice issuer's system screen.

[0025] The AI-OCR reading result identifies the "issuer" as "Saito Patent Office," but a search and query of the invoice recipient's customer master (a database storing information on the invoice recipient's customers; not shown in the diagram) reveals that "Saito Patent Office" does not exist in that customer master. In this case, the AI-OCR reading result is highly likely to be a misrecognition, so a warning mark 63 is displayed, and to its right, an explanation of the reason for the warning, "Not found in customer master," is displayed. This allows the person in charge at the invoice issuer to realize that the issuer's name has been misrecognized by the AI-OCR and to correct it to the correct name.

[0026] The input box on the left side of the display screen 60 allows for value modification. When the invoice issuer corrects the amount and issuer name to the correct values, and then places the mouse pointer (cursor) over the "Send" button 66 and clicks, the corrected text information and invoice image are sent to and saved in a database (323 in Figure 3) accessible to the invoice recipient (S15). Since the text information about the invoice saved in database 323 is the correct information modified by the invoice issuer, the invoice recipient does not need to verify it. This reduces the burden of verification on the invoice recipient.

[0027] The text information corrected by the invoice issuer, along with the content of the warnings notified to the invoice issuer and the revision history by the invoice issuer's staff, may be sent to the invoice recipient's database 323. This allows the invoice issuer's staff to be aware of the warning content and the revision history by the issuer's staff, reducing the burden on the invoice recipient's staff for verification.

[0028] In this embodiment, even when warning marks 62 and 63 are displayed, pressing the "Send" button 66 will send the invoice text information to the database 323. However, it is also possible to configure the system so that the "Send" button 66 cannot be pressed until the person in charge at the invoice issuer corrects the items with warning marks and all warning marks disappear.

[0029] Furthermore, a distinction could be made between cases that are definitely incorrect and cases that may be incorrect but are not. The former would display a warning mark indicating "high warning level," and the latter a warning mark indicating "low warning level." If even one "high warning level" remains, the "Send" button 66 cannot be pressed, but if there are no "high warning level" issues but some "low warning level" issues remain, the "Send" button 66 can be pressed.

[0030] Furthermore, if there are no errors in the amount in the invoice image, the issuer is listed in the invoice recipient's customer master, and no problems are detected in the AI-OCR reading results, the display screen 60 in Figure 6 will not be displayed. Instead, the text information of the invoice, which is the result of the AI-OCR reading, will be sent directly to the database 323. The display screen 60 in Figure 6 will only be displayed if a problem is detected in the AI-OCR reading results. This would reduce the burden on the invoice issuer.

[0031] Although embodiments and examples of the present invention have been described above, the technical scope of the present invention is not limited to the embodiments described above, and the above embodiments can be modified in various ways within the scope of the claims. [Explanation of symbols]

[0032] 11. Document Image Upload Section 12 Text Information Output Unit 13 Display Control Unit 14. Text Information Correction Reception Department 20 Computers 21 Memory section 22 Arithmetic section 23 Input section 24 Output section 31. Invoice issuer-side system 32. Invoice recipient-side system 311 Display section 312 Input section 321 External Uploader 322 AI-OCR 323 Databases 50 Upload screen 51 Upload Area 52 Send button 60 display screen 61 Invoice Image 62, 63 Warning Marks 64, 65 Warning Messages 66 Send button

Claims

1. The document image upload section uploads the document image, A text information output unit applies character recognition technology to the evidence image uploaded to the evidence image upload unit and outputs the text information recorded in the evidence image. A display control unit that causes the text information output by the text information output unit to be displayed on a display unit used by the issuer of the document, A text information correction receiving unit that accepts corrections to the text information displayed by the display unit, An accounting information processing system equipped with the following features.

2. The text information normality determination unit has a text information normality determination unit that outputs a determination result of whether the text information is normal or not based on the text information output by the text information output unit. If the display control unit determines that the text information is not normal, it will cause the display unit used by the issuer of the document to display that fact. The accounting information processing system according to claim 1.

3. The text information normality determination unit has a text information normality determination unit that outputs a determination result of whether the text information is normal or not based on the text information output by the text information output unit. If the text information normality determination unit determines that the text information is not normal, the display control unit causes the text information output by the text information output unit to be displayed on the display unit used by the issuer of the document. If the text information normality determination unit determines that the text information is normal, the display control unit transmits the document image to the recipient of the document image. The accounting information processing system according to claim 1.

4. The accounting information processing system according to claim 2, wherein the text information normality determination unit outputs a determination result of whether the text information is normal or not based on the text information relating to an amount output by the text information output unit.

5. The accounting information processing system according to claim 1, wherein the display control unit causes the display unit used by the issuer of the document to display a message to the effect that the text information relating to the name of the document issuer output by the text information output unit is not recorded in the recipient's customer master.

6. Executed by a computer, The document image upload step involves uploading an image of the document, The text information output step involves applying character recognition technology to the uploaded evidence image in the evidence image upload step and outputting the text information recorded in the evidence image. A display control step in which the text information output in the text information output step is displayed on a display unit used by the issuer of the document, The display control step includes a text information modification acceptance step that accepts modifications to the displayed text information, An accounting information processing method having the following characteristics.

7. On the computer, The document image upload step involves uploading an image of the document, The text information output step involves applying character recognition technology to the uploaded evidence image in the evidence image upload step and outputting the text information recorded in the evidence image. A display control step in which the text information output in the text information output step is displayed on a display unit used by the issuer of the document, The display control step includes a text information modification acceptance step that accepts modifications to the displayed text information, An accounting information processing program that executes this process.

Images