2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP) | 2019

On the evaluation of compact polarimetric SAR features over sea oil slicks during a controlled oil-on-water exercise

 
 
 
 
 

Abstract


Polarimetric Synthetic Aperture Radar (SAR) has powerful capability in marine oil spill detection since it could distinguish clean sea surface from mineral oil spill and false alarms as biogenic thin films. As an alternative to fully polarimetric (FP) SARs, which provide complete scattering information, compact polarimetric (CP) SARs transmit a polarized signal different from conventional horizontal and vertical ones, while receiving coherently the backscattered signal. Hence, they could obtain partial polarimetric information while maintaining a large swath-width, which makes it an ideal tool for marine oil spills observation. However, the proper selection of compact polarimetric SAR modes still remains a critical issue. By transmitting SAR signal with different polarizations, the scattering mechanism of the sea surface can be reflected by different CP SAR features. Hence, it is necessary to analyze the behavior of sea surface scattering as experienced by different CP SAR modes, i. e., different transmitted polarization. In this paper, a unified framework was applied to derive compact polarimetric SAR features under General Transmit and Linear Receive (GTLR) polarization configurations. Following the rationale of polarization signature, the characteristics of features derived by transmitting signals under different roll and ellipticity angles are analyzed. Experiments were conducted on a UAVSAR (Uninhabited Aerial Vehicle SAR) FP SAR image acquired at L-band during an oil-on-water exercise conducted offshore the Norwegian coast, where plant, crude oil, and oil emulsions with different oil-water mixtures were released at sea.

Volume None
Pages 1-4
DOI 10.1109/ICSIDP47821.2019.9173040
Language English
Journal 2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP)

Full Text