A fast sequential method for polygonal approximation of digitized curves
Computer Vision, Graphics, and Image Processing
Computational Perspectives on Map Generalization
Geoinformatica
A genetic algorithm approach to cartographic map generalisation
Computers in Industry - Special issue: Soft computing in industrial applications
| Hi-index | 0.00 |
Research on generalization implies a need for intensive experiments withthe tools developed. This need implies in turn that a research platform beavailable which allows for an easy development and integration of newalgorithms, and which provides interactive facilities for determining thebest tools sequences and tunings with respect to the geometricalcharacteristics of the features to generalize. We present in this paper sucha research platform, named PlaGe, which has been developed from 1992 on andis widely used for shareable research developments in our laboratory. Sofar, this platform has allowed the research team to implement most of theclassical algorithms as well as a technique for segmenting andcharacterizing line features. It is worth noting that this latter point isof prime importance in order to be able to apply the right processing withrespect to the local geometry. To automate further the generalizationprocess, starting with the currently available set of algortihms forsimplification and smoothing, requires the investigation of twocomplementary issues: devising algorithmical implementations for operatorssuch as caricature and schematization, which are often needed to cope withcomplex geometries, and determining a generalization strategy which, on thebasis of computed geometrical characteristics, results in the application ofthe best sequences. Four new algorithms, for caricature and schematization,are introduced in the following, as well as a generalization strategy. Theresults obtained so far indicate that we should be able to automatesignificantly the generalization process for practical production needs.