In this paper, we describe an extension of an automatic road extractionprocedure developed for single SAR images towards multiaspect SARimages. Extracted information from multi-aspect SAR images is notonly redundant and complementary, in some cases even contradictory. Hence, multi-aspect SAR images require a carefulselection within the fusion step. In this work, a fusion step based on probability theory is proposed. Before fusion, the uncertaintyof each extracted line segment is assessed by means of Bayesian probability theory. The assessment is performed on attribute-leveland is based on predefined probability density functions learned from training data. The prior probability varies with globalcontext. In the first part the fusion concept is introduced in atheoretical way. The importance of local context information and thebenefit of incorporating sensor geometry are discussed. The second part concentrates on the analysis of the uncertainty assessmentof the line segments. Finally, some intermediate resultsregarding the uncertainty assessment of the line segments using realSAR images are presented.     «

In this paper, we describe an extension of an automatic road extractionprocedure developed for single SAR images towards multiaspect SARimages. Extracted information from multi-aspect SAR images is notonly redundant and complementary, in some cases even contradictory. Hence, multi-aspect SAR images require a carefulselection within the fusion step. In this work, a fusion step based on probability theory is proposed. Before fusion, the uncertaintyof each extracted line segment is assessed by mean...     »