Document Image Masking Using Similar Item Name Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing apparatuses struggle with varying user-specific masking of regions on personal identification cards, requiring manual intervention to conceal sensitive information, which is cumbersome and prone to errors.

Innovation Solution

An image processing apparatus with an image reading portion, control portion, and storage portion, utilizing machine learning models and dictionary data to automatically detect and mask regions corresponding to specific personal information items, enabling user-defined masking without manual intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual masking intervention is used, then masking accuracy can be ensured, but operation complexity and time consumption increase

Engineering Contradiction:
Improvemasking accuracyVSAvoidoperation complexity
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system performs automatic item name recognition and region masking without requiring manual intervention. The control portion automatically detects item names in the document image, identifies their corresponding regions, and applies masking, enabling the system to serve itself rather than requiring user operation for each masking task

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical masking operations with automated image processing and character recognition systems. The control portion uses optical character recognition to identify item names and automatically determines masking regions, substituting human manual operations with computational processes

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If automatic masking is implemented, then operation time is reduced, but recognition accuracy may deteriorate

Engineering Contradiction:
Improveprocessing speedVSAvoidrecognition accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary character recognition and item name identification before the masking process. By pre-identifying all item names and their locations in the document, the system prepares the necessary information in advance, enabling accurate automatic masking without requiring manual intervention during the actual masking operation

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary character recognition process that bridges the gap between raw document images and accurate item identification. This intermediary step converts visual information into structured data that can be reliably processed for masking, improving the overall accuracy of the automatic system

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If user-specific masking is required, then data protection is improved, but device complexity increases

Engineering Contradiction:
Improvedata protectionVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system applies different masking treatments to different regions of the document based on their content type. Item names and their corresponding values are identified and masked separately, allowing selective protection of specific personal information regions while preserving other document content, thus providing localized data protection

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4687110A1Image processing apparatus
Publication Date: 2026.02.04 KYOCERA DOCUMENT SOLUTIONS INC
  • EP4687110A1 patent drawingFigure 1
  • EP4687110A1 patent drawingFigure 2
  • EP4687110A1 patent drawingFigure 3

AI summary

An image processing apparatus (100) includes an image reading portion (2), a control portion (10), and a storage portion (101). The storage portion (101) stores dictionary data (DD) in which, for one kind of item, a plurality of different similar item names are defined. The control portion (10) detects, from image data, a first region expected to individually include either a region of an item name or a region of an item value. For all the similar item names, the control portion (10) calculates their degree of character string similarity to a partial character string obtained by dividing the character string in the first region into substrings and, if the degree of character string similarity of any similar item name to the partial character string is equal to or higher than a threshold value, judges the first region as the region of the item name.