Efficient estimation of qualitative topological relations based on the weighted walkthroughs model

Weighted walkthroughs are a quantitative model for representing the spatial relation between two raster features in image databases. In this paper, we establish a correspondence between the weighted walkthroughs and qualitative models for spatial reasoning. We provide rules for estimating qualitative geometric properties and topological relations from the quantitative data that are computed for each pair of pixel sets. The approach has been tested through experiments with raster regions.