Jayant Sharma, Doug Flewelling, and
Max Egenhofer Sixth International Symposium on Spatial Data Handling, Edinburgh, Scotland, pp. 665-681, September 1994.
Abstract
Traditionally, GISs employ purely quantitative methods to represent
and infer spatial information. This approach has serious
shortcomings when dealing with qualitative spatial information,
which may be incomplete or imprecise and without knowledge of the
particular geometry of the spatial objects involved. This paper
describes efforts to build a prototype of a qualitative spatial
reasoner about spatial relations such as topological relations,
cardinal directions, and approximate distances. It builds on
relation algebras developed for the individual spatial relations.
The system is extensible as demonstrated by the inclusion of
temporal relations. The novel concept in this object-oriented
setting is the treatment of relations as first-class objects,
rather than as labeled links between spatial objects.