A model-based method is presented for recognition of roads and bridges in fully polarimetric Synthetic Aperture Radar (SAR) images. Polarimetric SAR features and geometric attributes of roads and bridges are used for segmentation, morphological filtering and recognition of roads and bridges. Roads and bridges are often segmented into small disconnected regions due to the presence of interfering objects. A series of Hough transformations are used to group the small regions, for recognition of bridges, a Constant False Alarm Rate (CFAR) detector is first used to extract strong backscatterers from bridge fences, the detected strong backscatterers are next grouped by a Hough transformation to find potential bridge fences, and bridges are then recognized using the polarimetric features of the regions between the potential fences. After recognition of bridges, a different Hough transformation is used to recognize roads. High-resolution SAR images acquired from MIT Lincoln Laboratory are used to illustrate our method.

Traditional constant false alarm rate (CFAR) detection algorithms produce a lot of false targets when applied to single-look, high-resolution, fully polarimetric synthetic aperture radar (SAR) images , due to the presence of speckle. We propose a two stage CFAR detector followed by conditional dilation for detecting point and extended targets in polarimetric SAR images. In the rst stage, possible targets are detected and false targets due to the speckle are removed by using global statistical parameters. In the second stage, the local statistical parameters are used to detect targets in regions adjacent to targets detected in the rst stage. Conditional dilation is then performed to recover target pixels lost in second stage CFAR detection. The performance of a CFAR detector will be degraded if an incorrect statistical model is adopted and the data are correlated. A goodness-of-t test is performed to decide the appropriate distribution and the eeects of decorrelation of the data are considered. Good experimental results are obtained when our method is applied to single-look, high-resolution, fully polarimetric SAR images acquired from MIT Lincoln Laboratory.

Traditional constant false alarm rate (CFAR) detection algorithms produce a lot of false targets when applied to single-look, high-resolution, fully polarimetric synthetic aperture radar (SAR) images, due to the presence of speckle. We propose a two stage CFAR detector followed by conditional dilation to improve CFAR detection algorithms. Good results are obtained when our method is applied to single-look, high-resolution, fully polarimetric SAR images acquired from MIT Lincoln Laboratory.< >