Spatial Information Theory

Volume 2205 of the series Lecture Notes in Computer Science pp 306-321


Recognition of Abstract Regions in Cartographic Maps

  • Joe Heike SteinhauerAffiliated withDepartment for Informatics, University of Hamburg
  • , Tom WieseAffiliated withDepartment for Informatics, University of Hamburg
  • , Christian FreksaAffiliated withDepartment for Informatics, University of Hamburg
  • , Thomas BarkowskyAffiliated withDepartment for Informatics, University of Hamburg

* Final gross prices may vary according to local VAT.

Get Access


In the human interpretation of cartographic maps the areas we shall call abstract regions consist of several symbols (map objects), which are grouped to a single object. This abstraction process is an important part of human map interpretation. Abstract regions exist as mental objects in a mental map in the interpreter’s mind. In this article we describe an approach to automate the human process of recognizing abstract regions in cartographic maps by technical processes. We designed and implemented a system for defining abstract regions by hierarchical descriptions. The hierarchies are represented by attributed grammars that can be translated by a compilercompiler to yield a parser for abstract regions. With this parser, abstract region candidates that were identified by simple rules can be evaluated to check if they conform to the definition provided by the user. Our approach combines cognitive considerations on human abstraction with techniques from theoretical computer science and artificial intelligence.


cognitive modeling formal methods map generalization map interpretation region abstraction spatial concepts