Silvio Dalfini

Learn More
In this paper a new automatic approach to road extraction from aerial images is proposed. This method improves a recently introduced promising approach to probabilistic contour tracking, originally semi-automatic, by adding a fully automatic initialization strategy and a merging methodology, able to combine the different obtained results. The initialization(More)
A complete system for extracting geographical entities from aerial images is proposed. The method treats separately the extraction of green areas, forestry and buildings from the extraction of roads. The former issue is addressed with a color segmentation followed by a region growing approach. The latter issue is addressed by a probabilistic contour(More)
The extraction and classification of geographical entities from aerial images has grown in importance in the last years, due to its wide applicability in different research areas. From the others, the GIS (Geographic Information System) topic is one of the most interesting, for both practical and theoretical motivations. In this context, annotation and(More)
  • 1