Defect-Aware Image Analysis Region Division for Infrastructure Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image-based inspection methods for infrastructure structures, such as bridges and tunnels, face inefficiencies due to the splitting of defects across multiple inspection regions, leading to misidentification and inefficient task distribution among workers.
Innovation Solution
An image processing apparatus that divides inspection regions based on defect status, including size, position, and concentration, to minimize defect splitting and improve efficiency by ensuring each defect is inspected within a single region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the inspection region is divided using a predetermined distance, then the inspection task is facilitated and worker efficiency is improved, but defects longer than the division size are split across multiple regions causing misidentification
Solution Approach 1:
The inspection region is divided into multiple sub-regions, but the division is intelligently controlled to ensure defects are not split. The system segments the region based on defect boundaries detected through image analysis, rather than simple geometric division.
Solution Approach 2:
The system performs preliminary image analysis to detect defects and determine their boundaries before dividing the inspection region. This preliminary action ensures that subsequent region division does not split defects, as the division boundaries are adjusted based on the detected defect locations.
2Ease of operation
If the inspection region is divided into multiple sections, then the inspection task becomes more manageable, but multiple workers checking different portions of the same defect lead to task inefficiency
Solution Approach 1:
The system provides feedback information to workers about which defects are located at region boundaries and may be visible in adjacent regions. This feedback mechanism prevents duplicate inspection of the same defect by multiple workers and ensures complete defect coverage.
Solution Approach 2:
The system acts as an intermediary that coordinates the inspection tasks among multiple workers. It manages the assignment of defects to specific workers and provides information about defect locations that span multiple regions, ensuring efficient task distribution.
3Shape
If defects are split across multiple division regions, then the region division achieves compact boundaries, but the overall defect image cannot be viewed in a single region requiring multiple checks
Solution Approach 1:
The system performs preliminary detection of defect boundaries and uses this information to adjust region division boundaries. This ensures that complete defect images are contained within single inspection regions, eliminating the need for workers to check multiple regions for the same defect.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
An image processing apparatus includes an image analysis unit (103) 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 (104) configured to divide a region subjected to the image analysis to acquire division regions, and the division unit (104) 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.