我要吃瓜

Conference Paper (unpublished)

A polarimetric change detector identifying power modifications of scattering mechanisms

Details

Citation

Marino A, Lopez-Sanchez JM & Hajnsek I (2012) A polarimetric change detector identifying power modifications of scattering mechanisms. 9th European Conference on Synthetic Aperture Radar, Nuremberg, Germany, 23.04.2012-26.04.2012.

Abstract
Monitoring land changes is an important topic in remote sensing. Polarimetric Persistent Scatterers (PPS) may increase substantially the number of candidate points for estimation of land movements. Beside the retrieval of PPS, in polarimetric and interferometric SAR (POLinSAR) the approximation of considering the scattering mechanisms between two acquisitions stable (Equi-Scattering Mechanisms, ESM) is largely employed. Consequently, an accurate and robust methodology is needed to identify appropriately the PPS and test the ESM fulfilment. In this paper, a new methodology exploiting polarimetric data is proposed to perform change detection. More in detail, the algorithm performs the optimisation of an error factor extracted from the POLinSAR coherence through a diagonalization. As a result, the maximum and minimum eigenvectors identify the scattering mechanisms suffering the largest and smallest change between the two acquisitions, respectively. Tests carried out with E-SAR DLR data (AGRISAR2006 campaign) are provided, showing the capability to identify targets that were moved between the two acquisitions

Keywords
Polarimeters; Radar; Change detection; Change detectors; Diagonalizations; Error factors; Interferometric SAR; Persistent scatterers; Polarimetric data; Scattering mechanisms

Journal
Proceedings of the European Conference on Synthetic Aperture Radar, EUSAR: Volume 2012-April

StatusUnpublished
Publication date31/12/2012
Conference9th European Conference on Synthetic Aperture Radar
Conference locationNuremberg, Germany
Dates

People (1)

Dr Armando Marino

Dr Armando Marino

Associate Professor, Biological and Environmental Sciences