S. Fortunati). ing (MRI), have highlighted the impulsive, heavy-tailed behaviour of the observations [1] . These experimental evidences have motivated the need to go beyond the Gaussian model and develop new statistical models able to better characterize the data. One of the more flexible and general non-Gaussian model is represented by the set of the Complex Elliptically Symmetric (CES) distributions [2] , also called Multivariate Elliptically Contoured distributions [3] . CES distributions encompasses the complex Gaussian, the Generalized Gaussian and all the Compound Gaussian (CG) distributions, such as the complex t -distribution and the K -distribution, as special cases. The pdf of a CES distributed N -dimensional random vector x l ∈ C N is completely characterized by the mean value γ, the scatter (or shape) matrix and by a real valued function w (t) : R + → R , called the density generator , i.e. x l ∼ CES N ( γ, , w ) [2,3] . The CES distributions have been used in a variety of applications, in particular in the radar and array signal processing fields.Other experimental evidences reveal recurring violations of the matched model assumption, that is the claim of a perfect match between the assumed and the true data model. The mathematical bases of a formal theory of the parameter estimation under model misspecification has been firstly developed by statisticians as Huber [4] , White [5] and Vuong [6] and recently rediscovered by the Signal Processing (SP) community [7-9] and applied to a va-
The satellite-borne SAR (Synthetic Aperture Radar) is a quite promising tool for high-resolution geo-surface measurement. Recently, there has been a great interest in Coherent Change Detection (CCD), where the coherence between two SAR images is evaluated and analyzed to detect surface changes. The sample coherence threshold may be used to distinguish between the changed and unchanged regions in the scene. Using COSMO-SkyMed (CSK) images, we show that for changed areas, the coherence is low but not completely lost. This situation, which is caused by the presence of bias in the coherence estimate, considerably degrades the performance of the sample threshold method. To overcome this problem, robust detection in inhomogeneous data must be considered. In this work, we propose the application and improvement of three techniques: Mean Level Detector (MLD), Ordered Statistic (OS) and Censored Mean Level Detector (CMLD), all applied to coherence in order to detect surface changes. The probabilities of detection and false alarm are estimated experimentally using high-resolution CSK images. We show that the proposed method, CMLD with incorporation of guard cells (GC) in the range direction, is robust and allows for nearly 4% higher detection probability in case of low false alarm probability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.