Building Change Detection via Stereo Photogrammetry and Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing building change detection methods are prone to erroneous detections due to influences from features other than buildings, such as trees, roads, and water bodies, which can lead to incorrect identification of building changes based on height variations.
Innovation Solution
A building change detection apparatus that acquires and compares image data over time to detect changes in features, excluding areas where buildings are absent, thereby accurately identifying changes specific to buildings by using stereo photogrammetry and image processing to differentiate between building and non-building features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If height data is extracted from photographed images to detect building changes, then automation is improved, but measurement precision deteriorates due to erroneous detections from non-building features
Solution Approach 1:
The patent segments the detection process into two distinct stages: first detecting all features with height changes using stereo photogrammetry, then separately identifying building absent areas using image processing. This segmentation allows the system to handle different types of features through different detection methods, improving overall precision while maintaining automation.
Solution Approach 2:
The patent extracts and excludes building absent areas from the detection target by using image processing to identify regions where no building exists. This extraction of problematic areas (trees, roads, water bodies) prevents them from causing erroneous detections, thereby improving measurement precision without reducing automation.
2Productivity
If all features are detected using stereo photogrammetry, then productivity is improved, but reliability deteriorates due to erroneous detections from trees, roads, and water bodies
Solution Approach 1:
The patent performs preliminary identification of building absent areas through image processing before final building change detection. By pre-identifying regions where no buildings exist (such as tree areas, roads, and water bodies), the system prevents erroneous detections in these areas, thereby improving reliability while maintaining the productivity gains from automated stereo photogrammetry.
Solution Approach 2:
The patent introduces image processing as an intermediary step between stereo photogrammetry and building change detection. This intermediary process identifies and marks building absent areas, allowing the main detection system to focus only on areas where buildings actually exist, thus improving reliability without sacrificing the automated productivity of the overall system.
3Measurement precision
If visual examination of all buildings is performed, then measurement precision is improved, but productivity deteriorates due to manual work requirements
Solution Approach 1:
The patent merges two different techniques: automated stereo photogrammetry for detecting height changes and image processing for identifying building absent areas. By combining these methods, the system achieves the precision of manual examination (through accurate building identification) while maintaining the productivity of automation (through automated height change detection), thus resolving the contradiction between precision and productivity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution effectively reduces erroneous detections by excluding non-building features from the analysis, allowing for more accurate detection of building changes, including new constructions and disappearances.
Implementation Method 1
extracts height data of a feature on the ground from a plurality of photographed images of the ground by stereo matching process
Data Source
AI summary
A building change detection apparatus detects a change in a building in a predetermined area on a map between a first time point and a second time point later than the first time point. A first image data acquiring unit acquires first image data of the predetermined area photographed at the first time point. A second image data acquiring unit acquires second image data of the predetermined area photographed at the second time point. A feature change detecting unit compares the first image data with the second image data to detect changes in features including a building in the predetermined area. A building absent area specifying unit specifies a building absent area in the predetermined area where a building is not present. A building change detecting unit detects a change in a building in the predetermined area by setting a feature whose change is detected by the feature change detecting unit in the predetermined area from which the building absent area is excluded, as a changed building in the predetermined area.