Detection of urban structures in SAR images by robust fuzzy clustering algorithms: The example of street tracking

In this work, we present a fuzzy approach to the analysis of airborne synth etic aperture radar (SAR) images of urban environments. In particular, we w ant to show how to implement structure extraction algorithms based on fuzzy clustering unsupervised approaches. To this aim, the idea is to segment first the sensed data and recognize ver y basic urban classes (vegetation, roads, and built areas). Then, from thes e classes, we extract structures and infrastructures of interest. The initi al clustering step is obtained by means of fuzzy logic concepts and the suc cessive analyses are able to exploit the corresponding fuzzy partition. As a possible complete procedure for urban SAR images, in this paper, we fo cus on the street tracking and extraction problem. Three road extraction al gorithms available in literature (namely, the connectivity weighted Hough t ransform (CWHT), the rotation Hough transform, and the shortest path extrac tion) have been modified to be consistent with the previously computed fuzz y clustering results. Their different capabilities are applied for the char acterization of streets with different width and shape. The whole approach is validated by the analysis of AIRSAR images of Los Ang eles, CA.