Cartographic Generalization of Line Elements based on Maximum Correlation
Nada Vučetić, Nedjeljko Frančula, Svetozar Petrović · Publication Database GFZ (GFZ German Research Centre for Geosciences) · 2006
Cartographic generalization is a special case of the generalization of information subjected to specific requirements. The present goal is full automa-tion, which is still a problem without a satisfactory solution, even for partial tasks, like the generalization of line features. Different algorithms for the latter can be found in the literature, probably all of them based on metric criteria. The basic idea of the approach proposed in this study (MCLS - Maximum Correla-tion Line Simplification) is to measure the similarity of forms using the correla-tion coefficient and to reduce the number of defining points in such a way that the simplified line remains as similar as possible to the original one. A solution of the problems of self-intersections and collisions with other objects for MCLS is presented as well. The advantages of the proposed new method are demonstrated by comparing it with the well-know Douglas-Peucker algorithm.