Inspection Support Image Recognition for Damage-Photo Organization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The manual organization of captured images and damage identification information during structure inspections is cumbersome and inefficient, particularly due to the complexity of associating large numbers of photographs with handwritten field notes.

Innovation Solution

An inspection support device and method that utilizes image recognition to automatically organize captured images and damage identification information by associating them with structural drawings and additional attributes, using a processor to recognize and link damage identification information with captured images through image data analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual organization of captured images and damage identification information is performed, then flexibility and adaptability are maintained, but productivity and ease of operation deteriorate due to cumbersome and inefficient work

Engineering Contradiction:

Improveorganization efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces manual mechanical organization processes with an automated image recognition system. The processor automatically recognizes damage identification information and captured image identification information from images, extracts relevant features, and associates them without human intervention, thereby eliminating the cumbersome manual sorting process while maintaining system flexibility.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs self-organization by automatically recognizing and associating damage information with captured images. The image recognition module autonomously extracts features, identifies relationships, and creates associations between damage identification information and captured image identification information, eliminating the need for manual sorting and organization work.

Inventive Principle:

Principle #25Self-service

2Loss of time

If manual association of photographs with field notes is performed, then accuracy can be maintained through human judgment, but loss of time increases due to complicated work

Engineering Contradiction:

Improveorganization timeVSAvoidassociation accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The patent replaces manual association processes with automated image recognition and feature extraction. The processor automatically recognizes damage identification information and captured image identification information, extracts relevant features, and associates them based on recognized patterns and relationships, achieving both speed and accuracy in the association process.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system incorporates feedback mechanisms where the image recognition module continuously validates and adjusts associations based on recognized patterns. The processor verifies the association between damage identification information and captured image identification information through feature matching and relationship validation, ensuring accuracy while reducing time consumption.

Inventive Principle:

Principle #23Feedback

3Productivity

If automated image recognition is used to organize captured images, then productivity improves, but device complexity increases

Engineering Contradiction:

Improveorganization speedVSAvoidprocessing system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the image processing system into distinct functional modules: an image acquisition module for capturing images, an image recognition module for extracting features and recognizing information, and a data association module for linking damage identification information with captured image identification information. This segmentation reduces overall system complexity by making each module independent and manageable while maintaining high productivity.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12449373B2Inspection support device, inspection support method, and program

Publication Date: 2025.10.21 FUJIFILM CORP

AI summary

In an inspection support device that organizes a captured image including damage to a structure and includes a processor, the processor acquires image data including information regarding a structural drawing of a target structure on a medium, and damage identification information regarding the damage and captured image identification information regarding the captured image of the target structure having the damage that are added by a user on the medium, recognizes the damage identification information through image recognition from the image data, recognizes the captured image identification information through image recognition from the image data, associates the damage identification information corresponding to a predetermined damage and the captured image identification information for the target structure having the predetermined damage, acquires the captured image corresponding to the captured image identification information, and associates the damage identification information and the captured image with each other.