Building Damage Extraction via Segmented Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating the soundness of buildings are inefficient, requiring extensive expertise and time, and struggle to accurately identify and prioritize damage in specific parts of infrastructure, leading to difficulties in rapid and accurate inspections due to budget, personnel, and geographical constraints.
Innovation Solution
A damage information extraction device that acquires structure information, specifies inspection positions and types of damage based on this information, and extracts relevant damage data from image data, allowing for focused inspections without needing to individually inspect all members, and includes a soundness determination unit to evaluate the building's soundness based on the extracted data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If comprehensive inspection of all building members is performed to ensure accurate soundness evaluation, then measurement precision is improved, but inspection time and resource requirements increase significantly
Solution Approach 1:
The building structure is divided into multiple members, and the inspection process is segmented into automated image capture and human expert evaluation phases. The automated system captures images of all members systematically, while human experts focus only on analyzing suspicious members identified by the system, thus reducing overall inspection time while maintaining accuracy.
Solution Approach 2:
An automated image capture and processing system acts as an intermediary between comprehensive data collection and expert evaluation. This intermediary system pre-processes the inspection by capturing and analyzing images to identify suspicious members, filtering out obviously sound members before human expert review, thereby reducing the time experts need to spend while preserving evaluation accuracy.
2Measurement precision
If expert technicians conduct detailed inspections to ensure accurate damage identification, then measurement precision is improved, but device complexity and operational difficulty increase
Solution Approach 1:
The automated inspection system performs self-service by automatically capturing images, identifying suspicious members, and generating inspection reports. This reduces the burden on expert technicians to perform routine data collection and initial analysis, allowing them to focus their expertise on critical evaluation of suspicious members, thereby simplifying their operational tasks while maintaining high accuracy.
Solution Approach 2:
Manual mechanical inspection processes are replaced with an automated digital image capture and processing system. The automated system handles image acquisition, preprocessing, and suspicious member identification, replacing the need for complex manual inspection procedures and reducing the operational complexity for expert technicians.
3Productivity
If automated image processing is used to reduce inspection time, then productivity is improved, but measurement precision may deteriorate due to loss of expert judgment
Solution Approach 1:
The inspection process is segmented into automated image capture and processing for efficiency, and expert evaluation for precision. The automated system handles high-volume data collection and preliminary analysis to improve productivity, while expert technicians focus their judgment on suspicious members identified by the automated system to ensure measurement precision is maintained.
Solution Approach 2:
The automated image processing system serves as an intermediary that prepares and filters data before expert evaluation. It captures and processes images systematically, identifies suspicious members, and presents only relevant cases to experts, thereby improving productivity while preserving expert judgment where it is most needed for accurate damage assessment.
Data Source
AI summary
An inspection specifying unit acquires structure information of a building that is an inspection target. Use environment history information of the building that is an inspection target is acquired. The inspection specifying unit specifies a member position that is an inspection target and a type of damage that is an inspection target at the member position on the basis of the structure information of the building and the use environment history information. An inspection data acquisition unit acquires inspection data including image data obtained by imaging a subject image of each member and position information of each member. An inspection data analysis unit acquires the image data corresponding to the member position from the inspection data acquisition unit, analyzes the image data, and determines whether there is damage of the damage type specified as the inspection target, and a degree of damage in a case where there is the damage.