Multi-baseline SAR Interferometry using Elaboration of Amplitude and Phase Data

Carmine Abbate,Annarita Evangelista,Roberto Di Folco

doi:10.13189/ujeee.2015.030204

Abstract

Multi-baseline interferometric Synthetic Aperture Radar (In-SAR) systems can be exploited to estimate the Digital Elevation Model (DEM) of the observed scene without ambiguities and with an increased accuracy, even in the case of high sloped ground regions. The techniques usually used exploit only the interferometric phase information and they are based on Maximum Likelihood (ML) estimation. An important problem to be taken into account is the mutual correlation of the (complex) interferometric images which impedes the closed form evaluation of the interferometric phases likelihood function. Moreover the statistical independence approximation of the phase interferograms is usually adopted. In this paper we present a method exploiting both amplitude and phase of the interferometric images, with the purpose of expressing the multi-baseline likelihood function in closed form, and we show that, when the number of baselines increases, to achieve an higher estimation accuracy the images mutual correlation cannot be neglected. We also show that to obtain a full resolution speckle reduced intensity image from several full resolution multi-baseline interferometric (complex) images, a phase compensation and a whitening operation have to be performed before averaging the data intensities.

Highlights

In this paper we present a method exploiting both amplitude and phase of the interferometric images, with the purpose of expressing the multi-baseline likelihood function in closed form, and we show that, when the number of baselines increases, to achieve an higher estimation accuracy the images mutual correlation cannot be neglected
Synthetic Aperture Radar Interferometric systems (In-SAR) allow the estimate of the Digital Elevation Models (DEM) of the observed scene [1, 2], starting from two SAR images of the same scene, acquired with slightly different view angles along two different trajectories, spatially separated by a distance called baseline
In this paper we presented a method to improve the DEM estimation accuracy, starting from correlated multi-baseline interferometric images and a low resolution inaccurate a priori DEM of the observed scene

Summary

Introduction

Synthetic Aperture Radar Interferometric systems (In-SAR) allow the estimate of the Digital Elevation Models (DEM) of the observed scene [1, 2], starting from two SAR images of the same scene, acquired with slightly different view angles along two different trajectories, spatially separated by a distance called baseline (single-baseline configuration). Closed-form the “exact” likelihood function of the multi-baseline data, properly taking into account the mutual correlation (coherence) of all the interferometric images. This expression can be derived from the well known multivariate Gaussian model of the interferometric (complex) images, obtained in the fully developed speckle assumption [18]. Numerical results on simulated data show that, when the number of the baselines increases, to obtain an improvement in the DEM reconstruction accuracy, it is crucial to properly take into account the images coherence They show that to obtain a full resolution speckle reduced amplitude image, both coherence and height profile have to be exploited in the multi-baseline data processing

Statistical Model of Multi-baseline Images

Maximum Likelihood Estimation of Height Profile and Reflectivity Amplitude

Numerical Results

Conclusions