Science Service System

Summary of Proposal MTH0183

TitleExploitation of fractal scattering models for TerraSAR images interpretation
Investigator Riccio, Daniele - Universitą di Napoli, Ingegneria Elettronica e delle Telecomunicazioni
Team Members
Professor Riccio, Daniele - Universitą di Napoli, Ingegneria Elettronica e delle Telecomunicazioni
Tupin, Florence - Telecom ParisTech,
SummaryThe proposal is aimed at exploiting the use of microwave fractal backscattering models to the analysis of TerraSAR data. This activity is worth to be explored in view of the following general considerations. Fractal models provide the best description for natural surface profile: this was assessed by several mathematicians in the last century. In the last decades geologists widely proved, motivated and verified this assumption. In the last decade the group that proposes this activity has developed many scattering models based on a fractal description of the surface. These models have been published in the open literature, and compared to classical scattering models. It has been verified that, whenever natural surface are in order, the developed fractal scattering models provide a better description of the scattering mechanism in comparison with classical scattering models. More specifically, TerraSAR will provide an unique high resolution-wide coverage data set, including scattering data at one meter of resolution and, in the future, polarimetric data. At this resolution scales the use of fractal models is most promising whenever compared to the resolution scales typical of existing SAR spaceborne sensors: this assumption was partially verified by employing SAR airborne high resolution-small coverage data. We propose to perform a retrieval of soil fractal roughness parameters and dielectric constant by comparing real and simulated SAR images, according to the procedure detailed below: 1. Some calibrated geocoded TerraSAR images will be acquired. The attention will be focused on some test sites were DEMs are available. 2. Fractal estimation of the DEM fractal parameters will be performed. 3. The microscopic roughness (at spatial scales of the order of the employed electromagnetic wavelength) will be estimated by means of the fBm and WM fractal models. 4. The backscattering from each relevant resolution cell will be computed. 5. The simulated backscattered field and the actual one will be compared in order validate the fractal scattering models. 6. The procedure will be iterated in order to enlighten any discrepancy between the estimated fractal parameters of the surface (including the microscopic scale) and the actual ones. Expected outcome of the project is the definition, validation and set up of the electromagnetic scattering fractal models to be employed to develop efficient and reliable interpretation and inversion tools well suited for TerraSAR-X data. These tools are the best candidates to provide, in a semi-automatic way, geometrical, value added, information about natural surfaces.

Back to list of proposals

© DLR 2004-2016