A Framework of Model-oriented Map Generalization and Its Implementation

Map generalization is an international problem in cartography, which strongly needs human decisions and intelligent reasons integrated with many other research fields. Nowadays there is still a long way to go to realize complete automatic map generalization. An efficient way is human-machine interactive map generalization. We can divide the process of generalization into two parts: iterative labors of computer and artificial decisions, which can complete the generalization tasks successfully. The compartmentalization of map generalization theory, that’s model generalization and cartographic generalization, provides the academic elements of interactive generalization. Model generalization is the foundation of map generalization, which modifies the internal representation of map objects, makes it simpler and more convenient for the task of generalization. Based on a separation of the topological, metric and semantic aspects of map objects the paper presents a circular framework supporting this task. Supported by the map generalization software “AutoMap”, we outline the theoretical framework and its implementation.