3D analysis with high-level primitives: A crystallographic approach

Abstract

This paper introduces a new approach to 3D handling of geographical information in the context of risk analysis. We propose to combine several geometrical and topological models for 3D data to take advantage from their respective capabilities. Besides, we adapt from crystallography a high-level description of geographical features that enables to compute several metric and cardinal relations, such as the "lay on" relation, which plays a key-role for geographical information. © 2006 Springer-Verlag Berlin Heidelberg.