Qualitative Shape Representation
Christoph Schlieder · 2020
This chapter discusses the problems that arise with undetermined boundaries from a spatial reasoning perspective. Spatial reasoning is a subfield of AI research on knowledge representation which studies formalisms for encoding spatial information. One major concern in this field is the comparison of different representational formalisms in terms of their expressive power; another is the analysis of the computational costs of the underlying inference mechanisms. The approach to shape representation applies the idea of qualitative abstraction which is implicitly used in many spatial reasoning formalisms to a special kind of spatial information, namely, ordering information. The chapter focuses on the indeterminacy of spatial information in general and describes the procedure of qualitative abstraction and the concept of ordering information. It then develops the ordering information approach to qualitative shape description. By combining qualitative abstraction and ordering information a whole class of formalisms for reasoning about shape may be obtained.