Forest structure characterization using SAR tomography and an adaptive estimation technique

Conference: EUSAR 2022 - 14th European Conference on Synthetic Aperture Radar
07/25/2022 - 07/27/2022 at Leipzig, Germany

Proceedings: EUSAR 2022

Pages: 4Language: englishTyp: PDF

Authors:
Ferro-Famil, Laurent (ISAE-SUPAERO, University of Toulouse, France & CESBIO, CNES/CNRS/INRAE/IRD/UPS, University of Toulouse, France)
Huang, Yue (IETR, University of Rennes 1, France)
Ge, Nan (Technical Univerity of Munich, Germany)

Abstract:
3-D SAR imaging through tomographic focusing techniques offers a way to remotely characterize the structure of forests and estimate some of their bio- and geo-physical features. Stacks of Multi-Baseline InSAR images being generally composed of a small number of items, imaging results, based on the coherent combination of the different elements of the data set, may be affected by processing artifacts and limited vertical resolution. Following the original idea of Aguilera et al. [1], based on the use of a wavelet basis in order to represent a continuous reflectivity profile with a few coefficients, this paper proposes a rigorous derivation of a parametric Maximum Likelihood estimator vertical reflectivity profiles. This techniques uses a basis of functions which can be generated by a wavelet operator, and proposes an automated and fast optimization technique in order to estimate deconvolved and enhanced reflectivity features, that may be used analyze the structure of forests. An application to tomographic data acquired over a tropical forest at P band permits to assess the performance of this method.