Image Analysis Region Division That Avoids Splitting Defects

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image-based inspection methods for infrastructure structures divide inspection regions in a way that can split defects, leading to inefficiencies and misidentification of defects, especially when multiple workers are involved.

Innovation Solution

An image processing apparatus that divides inspection regions based on the status of defects, including size, position, and type, to minimize defect splitting and improve efficiency by using a division unit that considers defect surrounding regions and severity levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If the inspection region is divided using a predetermined distance, then the inspection task is facilitated by displaying each section, but defects longer than the division size are split across multiple regions leading to misidentification

Engineering Contradiction:

Improveinspection task facilitationVSAvoiddefect length measurement

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The inspection region is divided into multiple sub-regions, but the division is performed adaptively based on defect characteristics. The division unit determines division positions by analyzing defect length and concentration, ensuring that division boundaries do not cut through defects. This segmentation approach maintains ease of operation while preventing defect splitting.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different regions are divided with different characteristics based on local defect conditions. Regions with long defects or high defect concentration use larger division sizes or adjusted division positions, while other regions use standard division. This local adaptation ensures measurement precision is maintained where needed while preserving operational efficiency overall.

Inventive Principle:

Principle #3Local quality

2Productivity

If multiple workers share inspection tasks across different division regions, then task distribution is improved, but defects split across regions cause inefficiency and degradation underestimation

Engineering Contradiction:

Improvetask distribution efficiencyVSAvoiddefect inspection accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

Before distributing inspection tasks to multiple workers, the system performs preliminary analysis to identify defect positions, lengths, and concentrations. Based on this analysis, the division unit pre-determines optimal division positions that prevent defect splitting. This preliminary action ensures that subsequent parallel inspection by multiple workers maintains both productivity and reliability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from defect analysis results to adjust division strategies. When defects are detected, their characteristics feed back into the division unit's decision-making process, which then adjusts division positions and sizes accordingly. This feedback mechanism ensures that task distribution to multiple workers does not compromise inspection accuracy.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20260030732A1Image processing apparatus, image processing method, and non-transitory computer-readable storage medium

Publication Date: 2026.01.29 CANON KK

AI summary

An image processing apparatus includes an image analysis unit configured to perform image analysis on an image to be inspected to acquire defects obtained as a result of the image analysis and a division unit configured to divide a region subjected to the image analysis to acquire division regions, and the division unit divides the region subjected to the image analysis on the basis of a status of the defects. The status of the defects includes at least the size and position of each defect.