Oil spill identification using SVM based on polarization parameters

Abstract

Abstract: Due to the fact that oil spill has been increasing damage to the marine environment, to quickly detect the oil spill is certainly of significance for preventing and alleviating the disasters. Currently using polarized SAR in the detection of the oil spill is an important approach of remote sensing. In this paper, the SIR-C data were employed in the monitoring of oil spills, in which the polarized parameters entropy, scattering angle and anti-entropy were first extracted and then used to retrieve the oil spill information using the support vector machine (SVM) classifier. The assessment showed that SVM-based classification can achieve more accurate result than that of H-α decomposition based oil spill identification of SAR data.

Key words: polarization SAR, oil spill identification, SVM, test

CLC Number:

TN957.52
X55

ZOU Ya-rong, LIANG Chao, ZENG Tao. Oil spill identification using SVM based on polarization parameters[J]. Journal of Marine Sciences, 2013, 31(3): 71-75.

References

[1] WRIGHT J W. Backscattering from capillary waves with application to sea clutter[J]. IEEE Transactions on Antennas and Propagation,1966,14(6):749-754.
[2] FERRARO G, BASCHEK B, DE MONTPELLIER G, et al. On the SAR derived alert in the detection of oil spills according to the analysis of the EGEMP[J]. Marine Pollution Bulletin,2010,60(1):91-102.
[3] FORTUNY-GUASCH J. Improved oil slick detection and classification with polarimetric SAR[C]//Proceedings of the Workshop on POLinSAR: Applications of SAR Polarimetry and Polarimetric Interferometry,14-16 January 2003,Frascati, Italy (ESA SP-529),2003.
[4] MIGLIACCIO M, TRANFAGLIA M. A study on the capability of SAR polarimetry to observe oil spills[C]//Proceedings of the POL inSAR 2005 Workshop,17-21 January 2005,ESRIN, Frascati, Italy(ESA SP-586),2005:25.
[5] VELOTTO D, MIGLIACCIO M, NUNZIATA F, et al. Dual-polarized Terrasar-x data for oil-spill observation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011,49(12):4 751-4 762.
[6] MIGLIACCIO M, GAMBARDELLA A, MASSIMO T. Oil spill observation by means of polarimetric SAR data[C]//Proceedings of SEASAR 2006, 23-26 January 2006, Frascati, Italy(ESA SP-613),2006.
[7] NUNZIATA F, GAMBARDELLA A, MIGLIACCIO M. Dual-polarized SAR data for oil spill detection[C]//Proc. ‘Envisat Symposium 2007’, Montreux, Switzerland,23-27 April 2007 (ESA SP-636),2007.
[8] MIGLIACCIO M, NUNZIATA F. GAMBARDELLA A. On the co-polarized phase difference for oil spill observation[J]. International Journal of Remote Sensing,2009,30(6):1 587-1 602.
[9] NUNZIATA F, GAMBARDELLA A. MIGLIACCIO M. On The mueller scattering matrix for SAR sea oil slick observation[J]. IEEE Geoscience and Remote Sensing Letters,2008,5(4):691-695.
[10] CLOUD S R. Polarimetry: The characterization of polarization effects in EM scattering[D]. Birmingham: Birmingham University,1986.
[11] WANG Chao,ZHANG Hong,CHEN Xi, et al. Full polarization synthetic aperture radar image processing[M]. Beijing: Science Press,2008:107-126.
王超,张红,陈曦,等. 全极化合成孔径雷达图像处理[M]. 北京:科学出版社,2008:107-126.