In maritime applications involving estimation of radar sea clutter properties, non-sea-clutter targets and transitions between statistically different oceanographic conditions in the estimation window may lead to inaccurate modeling. Referring to mixtures in the estimation window as contamination, this work introduces a novel sea clutter contamination test based on log-cumulants from Mellin kind statistics [1]. It measures the significant deviation in log-cumulant space due to the contamination, and appears to be an effective tool for improving the sea clutter estimation or to be a direct first-stage target detector. The proposed contamination test is examined with real single look complex (SLC) fine resolution quad-polarimetric Radarsat-2 synthetic aperture radar (SAR) measurements, from the Norwegian Sea, under various oceanographic conditions.

Date Published: 21 November 2012
PDF: 14 pages
Proc. SPIE 8536, SAR Image Analysis, Modeling, and Techniques XII, 85360I (21 November 2012); doi: 10.1117/12.976879

Published in SPIE Proceedings Vol. 8536:
SAR Image Analysis, Modeling, and Techniques XII
Claudia Notarnicola; Simonetta Paloscia; Nazzareno Pierdicca, Editor(s)