Title :
Quality assessment of despeckled SAR images
Author :
Dellepiane, Silvana G. ; Angiati, Elena
Author_Institution :
Dept. of Biophys. & Electron. Eng. (DIBE), Univ. of Genoa, Genova, Italy
Abstract :
An innovative approach is here proposed for the quality assessment of despeckled SAR images. Both statistic and frequency analysis have been developed. Different statistic indexes, presented in literature and based on mean and standard deviation values, have been considered and new indexes are then proposed, able to gain a quantitative evidence of the visual comparison. The novel frequency analysis here presented allows to measure quantitatively mean preservation and isotropic behaviour of filters and overcomes the problems of the deterministic and statistical approaches. Different non-linear adaptive filters have been tested on various Cosmo/Skymed images acquired in different configuration modes.
Keywords :
adaptive filters; nonlinear filters; radar imaging; statistical analysis; synthetic aperture radar; Cosmo/Skymed images; despeckled SAR images; deterministic approach; frequency analysis; isotropic behaviour; nonlinear adaptive filters; quality assessment; standard deviation values; statistic indexes; statistical approach; Adaptive filters; Digital filters; Indexes; Maximum likelihood detection; Nonlinear filters; Speckle; Statistical analysis; SAR; de-speckling; frequency domain; quality assessment; statistic;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2011 IEEE International
Conference_Location :
Vancouver, BC
Print_ISBN :
978-1-4577-1003-2
DOI :
10.1109/IGARSS.2011.6050059