Abstract
Fuzzy clustering algorithms are used for the interpretation of high resolution SAR images of urban environments. The idea is to define a pyramidal procedure suitable for the characterization first of more different environments (for instance, green areas, streets, and buildings). This rough analysis is then followed by more oriented fuzzy clustering tools, devoted to the extraction of more details: in this paper a modified fuzzy Bough transform is introduced and used to group pixels classified as pixels in consistent straight lines.

This publication has 5 references indexed in Scilit: