A new geospatial overlay method for the analysis and visualization of spatial change patterns using object-oriented...

Area of Science:

• Geographic Information Systems (GIS)
• Remote Sensing
• Spatial Analysis

Background:

• Traditional GIS overlay methods often result in complex, multi-step processing for change detection.
• Calculating topology on-the-fly for polygon layers leads to intricate intersection results requiring further aggregation.
• Existing workflows lack inherent topological support, complicating meaningful data analysis.

Purpose of the Study:

• To present a novel, automated geospatial overlay method within a topologically enabled framework.
• To overcome the limitations of traditional GIS overlay routines in handling complex polygon intersections.
• To enable efficient and aggregated change detection analysis.

Main Methods:

• Implementation of a multi-scale vector/raster data model in eCognition software.
• Utilizing object-based image analysis with inherent topological relationships for object-by-object comparison.
• Leveraging classification, generalization, and aggregation concepts from object-oriented data modeling.

Main Results:

• Automated generation of aggregated change-detection layers.
• Interactive definition of change classes and dependencies via class hierarchy and inheritance.
• Successful application demonstrated on CORINE Land Cover datasets for change comparison.

Conclusions:

• The developed method offers a flexible, transferable, and fully automated solution for geospatial change detection.
• Object-based approaches with inherent topology significantly streamline complex spatial analysis.
• This framework enhances the efficiency and interpretability of land cover change studies.