Fractal Models for Sar Images of Natural Surfaces

In this paper an innovative model that describes the SAR imaging process of natural scenes is presented. It is completely fractal-based. In the hypothesis of small slope regime of the surface, we evaluate here, in closed form, the Power Spectrum Density of the SAR image. The proposed model effectively accounts for the effect of finite sensor resolutions and for the peculiarity of SAR acquisition geometry. Inverting this complete direct model it is possible to retrieve the fractal dimension of the observed surface. Significant experimental results, that validate the theoretical conclusions, are also shown.