23 October 2014 Robust discrimination of permanent scatterers using Cameron Decomposition
Author Affiliations +
Abstract
One of the main difficulties encountered in Differential Interferometry (DInSAR) applications is temporal and spatial decorrelation over time. Single pixels, called Permanent Scatterers (PS), overcome this difficulty since they are coherent over time and over wide look-angle variations. Permanent Scatterers identification using interferographic techniques is unfeasible since they require the use of many acquisitions. Samsonov and Tiampo have presented a technique that selects Permanent Scatterers by analyzing their Polarization Phase Difference (PPD). The PPD approach would work just fine looking for single bounce scatterers because they are invariant to any initial arbitrary rotation between the scatterer and the radar Line of Sight (LOS). We propose to replace the PPD technique with Cameron’s Coherent Target Decomposition (CTD) because it is more accurate in finding the single and double bounce scatterers as it eliminates the initial orientation angle of the scatterer. Additionally, Cameron’s CTD is capable of recognizing more scattering mechanisms which means that more pixels, depending on their amplitude and stability over time, can be classified as Permanent Scatterers. A sample scene of fully polarimetric SAR image depicting the San Francisco bay was employed for experimentation. Our results demonstrate the superiority of the Cameron's CTD approach compared to PPD’s approach for the selection of pixels classified as Permanent Scatterers.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
G. Kouroupis, P. Elias, V. Anastassopoulos, "Robust discrimination of permanent scatterers using Cameron Decomposition", Proc. SPIE 9245, Earth Resources and Environmental Remote Sensing/GIS Applications V, 92450M (23 October 2014); doi: 10.1117/12.2067216; https://doi.org/10.1117/12.2067216
PROCEEDINGS
8 PAGES


SHARE
Back to Top