Spatial Clustering of Related Structured Objects for Topographic Map Interpretation

In topographic map interpretation, spatial clustering, i.e., detecting a mosaic of nearly homogeneous sites, is a means for automatically detecting the continuity of morphological environments characterizing a map area. The goal is to group structured objects, i.e. spatial data collected at different sites, such that data inside each cluster models the continuity of morphologic or socio-economic environment, while separate clusters model variation over the space. Continuity is evaluated according to the discrete spatial structure that is (spatial) relations between separate sites implicitly defined by their geometrical representation and positioning. Only data collected within sites that are transitively connected in the discrete spatial structure can be clustered together according to a similarity judgment. The similarity among structured objects is computed as degree of matching with respect to a common generalization. Strong points of the proposed technique for this task are the power of first-order formalism as a knowledge representation and reasoning means and the capability of discovering spatial clusters which model the continuity of some spatial phenomena. The technique has been implemented in CORSO and tested in the interpretation of the topographic map of an area of the Apulia region in Italy.