A fuzzy-based method for road area detection in high resolution SAR images


Department of Geomatics Engineering, College of Engineering, Tehran University, Tehran, P.O. Box 11365-4563, Iran


Automatic road extraction from satellite images is one of the most important areas of research in the eld of remote sensing. The method proposed in this study is based on a fuzzy method for the detection of road areas from high resolution SAR images. In this method, the multiple features are extracted rst, using the backscatter coecients of each pixel and its neighboring pixels. The extracted features are combined with each other in the next step using a fuzzy algorithm, and, nally, the desired road areas are selected separately considering spatial and spectral criteria. The proposed algorithm is tested on di erent scenes of TerraSAR-X images. Experimental results reveal that the proposed method is e ective.