Document Image Comparison for Automatic Digitization Area Setting

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing reading systems require manual user input to set areas for digitization, which is cumbersome when dealing with large numbers of areas, leading to reduced operability.

Innovation Solution

An information processing apparatus that automatically extracts candidate areas for digitization by comparing unwritten and written documents, performing character and shape recognition, and setting these areas based on recognition results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual user input is used to set areas for digitization, then the system can accurately identify user intentions, but the operability deteriorates when dealing with large numbers of areas

Engineering Contradiction:

ImproveoperabilityVSAvoidcomplexity of area setting process

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system automatically extracts candidate areas for digitization by comparing unwritten and written documents, eliminating the need for manual user input to define areas. The automated process uses image data comparison and recognition algorithms to independently identify and set areas, thereby significantly improving operability while maintaining accuracy in handling large numbers of areas.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical area selection with automated image processing mechanisms. By using image data comparison between unwritten and written documents, along with character and shape recognition processes, the system substitutes human manual operation with automated computational methods to identify and set areas for digitization.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If automated area extraction is implemented, then operability improves, but measurement precision may deteriorate due to automatic recognition errors

Engineering Contradiction:

Improveease of area settingVSAvoidaccuracy of area identification

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system performs character recognition and shape recognition processes on extracted candidate areas to verify and refine the identification results. This feedback mechanism allows the system to adjust and correct automatic recognition errors, ensuring high measurement precision while maintaining the ease of automated operation.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system preliminarily extracts candidate areas through image data comparison, then subsequently performs recognition processes to verify accuracy. This two-stage approach ensures that both ease of operation and measurement precision are achieved by first providing automated extraction and then validating through recognition feedback.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260038293A1Information processing apparatus, information processing method, and non-transitory recording medium

Publication Date: 2026.02.05 PFU LTD

AI summary

An information processing apparatus includes circuitry that extracts one or more candidate areas to be digitized, based on a difference between first image data generated based on an unwritten document and second image data generated based on a written document; executes a character recognition process or a shape recognition process on the second image data to obtain a recognition result; and sets one or more areas to be digitized based on the candidate areas to be digitized and the recognition result.