# US 12,769,929 B2 · Information processing apparatus, non-transitory computer readable medium, and method for defining an arithmetic expression using feature and arrangement information

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,769,929 B2
- Document ID: US12769929B2
- Status: Granted
- Kind: B2
- Application number: 17/891,796
- Filed: 2022-08-19
- Published: 2023-09-28
- Granted: 2026-10-20
- Earliest claimed: 2022-03-22
- Assignees: FUJIFILM Business Innovation Corp.
- Applicants: Fujifilm Business Innovation Corp.
- Inventors: Miho Ishizuka
- Web page: https://patentgrep.com/patent/US12769929B2

## CPC

- G06F 40/174: Form filling; Merging
- G06F 17/11: for solving equations {, e.g. nonlinear equations, general mathematical optimization problems (optimization specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- G06V 30/19173: {Classification techniques}
- G06V 30/28: specially adapted to the type of the alphabet, e.g. Latin alphabet
- G06V 30/412: Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

An information processing apparatus includes a processor configured to: read plural pieces of character string information written on a form; obtain feature information indicating a feature relating to an arithmetic operation in which numerical information included in the plural pieces of character string information is used and arrangement information indicating a positional relationship between the plural pieces of character string information; and define, on a basis of the feature information and the arrangement information, an arithmetic expression for performing an arithmetic operation using an operator relating to the plural pieces of character string information.

## Claims

### Claim 1

An information processing apparatus comprising: 1.
a memory; and
a processor configured to:
obtain image data on which optical character recognition (OCR) processing is performed;
identify two set strings each comprising a unit and numerical information, located in adjacent columns within a same row of a table included in the image data;
extract the unit from each of the two set strings by parsing characters from the respective set string:
classify the extracted unit of each of the two set strings into a category by comparing the extracted unit to a category table stored in the memory;
determine an operator corresponding to the identified two set strings by referring to a combination of the categories of the two set strings in a prediction table stored in the memory; and
provide an arithmetic expression using the operator applicable to the two set strings in the same row of the table.

### Claim 2 (depends on 1)

The information processing apparatus according to claim 1, wherein the unit indicates at least quantity, amount of money, weight, or length.

### Claim 3 (depends on 2)

The information processing apparatus according to claim 2,
wherein, in the prediction table, the operation between a left-side term and a right-side term is specified in advance as a predicter operator.

### Claim 4 (depends on 3)

The information processing apparatus according to claim 3,
wherein the arithmetic expression includes an operation term in which the operator is used between numerical information of the two set strings and, as an operation result, numerical information included in a set string other than the two set strings.

### Claim 5

A non-transitory computer readable medium storing a program causing a computer to execute a process for processing information, the process comprising:
obtaining image data on which optical character recognition (OCR) processing is performed;
identifying two set strings each comprising a unit and numerical information, located in adjacent columns within a same row of a table included in the image data;
extracting the unit from each of the two set strings by parsing characters from the respective set string:
classifying the extracted unit of each of the two set strings into a category by comparing the extracted unit to a category table stored in a memory;
determining an operator corresponding to the identified two set strings by referring to a combination of the categories of the two set strings in a prediction table stored in a memory; and
providing an arithmetic expression using the operator applicable to the two set strings in the same row of the table.

### Claim 6

A method for processing information, the method comprising:
obtaining image data on which optical character recognition (OCR) processing is performed;
identifying two set strings each comprising a unit and numerical information, located in adjacent columns within a same row of a table included in the image data;
extracting the unit from each of the two set strings by parsing characters from the respective set string:
classifying the extracted unit of each of the two set strings into a category by comparing the extracted unit to a category table stored in a memory;
determining an operator corresponding to the identified two set strings by referring to a combination of the categories of the two set strings in a prediction table stored in a memory; and
providing an arithmetic expression using the operator applicable to the two set strings in the same row of the table.