Context-Aware Character Recognition Verification Interface

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Character recognition errors in image processing require manual confirmation and proofreading, which are time-consuming and burdensome for workers, despite existing UI methods that list and display similar character images for visual confirmation.

Innovation Solution

An information processing apparatus that extracts character strings and unit characters from document images, generates context information, and displays the extracted characters along with their context on a UI screen, facilitating efficient confirmation of character recognition results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual confirmation and proofreading operations are performed on extracted character results, then character recognition accuracy is improved, but time consumption and worker burden increase

Engineering Contradiction:
Improvecharacter recognition accuracyVSAvoidconfirmation operation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Context information acts as an intermediary element between the character image and the user. By displaying context information (such as surrounding characters, word boundaries, or semantic information) alongside the character image, the system provides additional cues that help users verify recognition accuracy more quickly without requiring full manual proofreading of each character.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system provides feedback to users by displaying context information that highlights potential recognition issues or confirms correct recognition. This feedback mechanism allows users to make faster judgment decisions about whether a character recognition is correct, reducing the time needed for manual confirmation while maintaining accuracy.

Inventive Principle:
Principle #23Feedback

2Productivity

If character images are listed and displayed for visual confirmation, then confirmation efficiency is improved, but lack of contextual information limits verification accuracy

Engineering Contradiction:
Improveconfirmation efficiencyVSAvoidcontextual information
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The system merges the character image display with context information display into a unified interface. Instead of showing only isolated character images, the invention combines them with surrounding textual context, making both the character image and its context simultaneously visible to the user, thereby enabling efficient verification without information loss.

Inventive Principle:
Principle #5Merging (Combining)

3Ease of operation

If isolated character images are displayed for confirmation, then ease of visual checking is improved, but absence of context reduces verification reliability

Engineering Contradiction:
Improvevisual checking easeVSAvoidverification reliability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system adds another dimension to the display by incorporating context information alongside the character image. This dimensional expansion transforms the display from showing only isolated characters to showing characters within their contextual framework, enabling users to maintain easy visual checking while simultaneously improving verification reliability through additional contextual cues.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS20250014372A1Information processing apparatus, method of controlling information processing apparatus, and storage medium
Publication Date: 2025.01.09 CANON KK
  • US20250014372A1 patent drawing
  • US20250014372A1 patent drawing
  • US20250014372A1 patent drawing

AI summary

An information processing apparatus includes: at least one memory that stores instructions; and at least one processor that executes the instructions to: extract a character string and a unit character included in the character string based on a result of character recognition processing on a document image; generate context information related to the unit character based on the extracted character string and attribute information indicating an attribute of the character string; and display the extracted unit character and the context information corresponding to the unit character on a UI screen.