Sea Oil Spill Detection Method Using SAR Imagery Combined with Object-Based Image Analysis and Fuzzy Logic

Abstract:

Synthetic aperture radar (SAR), a sensor with all weather and day and night working capacity, has been considered one of the most powerful tools for sea surface oil spill detection. However, lookalikes frequently appear in SAR images, limiting the operational use of SAR to detect oil spilled at sea. 20 scenes of Envisat ASAR images, which were acquired during the oil spill accident in the Gulf of Mexico in 2010, are utilized, with the objective to study how to better differentiate oil spills from lookalikes. 145 and 134 samples for oil spill and lookalike, respectively, are extracted, and their object-based geometric, physical and textural features are analyzed, in order to find the most effective features for oil spill classification. Based on the results of feature analysis, fuzzy logic (FL) is employed to construct a classifier for oil spill detection. One advantage of the proposed method is that it can produce the crisp probability of a dark segment being oil spill. The experiment shows that our method can derive promising result.

• Periodical: Advanced Materials Research (Volumes 1065-1069)
• Pages: 3192-3200
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.1065-1069.3192
• Online since: December 2014
• Authors: Teng Fei Su, Hong Yu Li*, Ting Xi Liu
• Keywords: Feature Analysis, Fuzzy Logic (FL), Object-Based Image Analysis, Oil Spill Classification, Synthetic Aperture Radar (SAR)

References

[1] Information on http: /www. coi. gov. cn/gongbao/zaihai.

[2] Information on http: /www. noaa. gov/deepwaterhorizon.

[3] Information on http: /en. wikipedia. org/wiki/Deepwater_Horizon_oil_spill.

[4] L. Shi: Sea Surface Oil Spill Detection Method by SAR and MODIS (Doctor dissertation, Ocean University of China 2008).

[5] C. Brekke, A. Solberg: Oil Spill Detection by Satellite Remote Sensing, Remote Sensing of Environment, Vol. 95 (2005) pp.1-13.

DOI: 10.1016/j.rse.2004.11.015

[6] W. Alpers, H. Huhnerfuss: Radar Signatures of Oil Films Floating on the Sea Surface and the Marangoni Effect, Journal of Geophysical Research, Vol. 93 (1988) pp.3642-3648.

DOI: 10.1029/jc093ic04p03642

[7] G. Fanny, G. Mercier, F. Collard, et al: Operational Oil-slick Characterization by SAR Imagery and Synergistic Data, IEEE Journal of Oceanic Engineering, Vol. 30 (2005) pp.487-495.

DOI: 10.1109/joe.2005.857526

[8] X. Liang, J. Zhang, J. Meng: Selection of Texture Characteristics in Classifying Oil Spill SAR Images, Advances in Marine Science, Vol. 25 (2007) pp.346-356.

[9] L. Chang, Z. Tang, S. Chang, et al: A Region-based GLRT Detection of Oil Spills in SAR Images, Pattern Recognition Letters, Vol. 29 (2008) p.1915-(1923).

DOI: 10.1016/j.patrec.2008.05.022

[10] H. Zhou, P. Chen: Oil Spill Identification in SAR Image Using Mahalobis Distance, Computer Engineering and Applications, Vol. 47 (2011) pp.195-203.

[11] H. Qu, H. Bo, W. Zhang: SAR Image Oil Spill Detection Based on Texture Feature and Support Vector Machine, Information Technology, Vol. 19 (2010) pp.90-93.

[12] T. Su: A Segmentation Algorithm for Oil Spill SAR Image Based on Hierarchical Agglomerative Clustering, Advances in Marine Science, Vol. 31 (2013) pp.256-265.