Coherence estimation of interferometric SAR images

Abstract

Usual coherence estimation in SAR interferometry is a time consuming task since an accurate estimation of the local frequency of the interferometrie fringes is required. In this paper a fast algorithm for generating coherence maps, mainly intended to data browsing, is presented. The proposed estimator is based on the speckle similarity of coherent SAR data and is, thus, independent of the fringes frequency. Advantages with respect to the usual estimates are achieved in terms of computational costs (up to 100 times lower), robustness (the estimator presented is not affected by possible local frequency estimation errors) and flexibility (the estimator can be applied both to complex and to detected images). The statistical properties of the frequency independent estimator are given in the stationary case. A preprocessing technique that reduces the degradions due to non-stationarities is then shown.