Oil spills reduce water surface roughness and can be detected by the Normalized Radar Cross-Section (NRCS) on SAR (Synthetic Aperture Radar) images where they appear as dark areas. With the purpose to detect oil slicks, in the last years a probabilistic method to distinguish oil spills from other similar oceanic features in marine (SAR) images has been developed and tested. The method uses statistical information obtained from previous measurements of physical and geometrical characteristics for both oil spill and natural features. A sample image is evaluated using a procedure to determine the probability that it is an oil spill. The classification-algorithm performance was evaluated using a test dataset of SAR images containing hundreds of examples of oil spills and of features exhibiting characteristics similar to oil spills (look-alike): more than 80% of the samples were classified correctly. The reliability of the method was then determined using a new dataset and similar results were obtained. The developed methodology and its capability in recognizing oil spills among look-alike are illustrated.
A method to detect oil spill based on SAR images
TRIVERO, Paolo
2002-01-01
Abstract
Oil spills reduce water surface roughness and can be detected by the Normalized Radar Cross-Section (NRCS) on SAR (Synthetic Aperture Radar) images where they appear as dark areas. With the purpose to detect oil slicks, in the last years a probabilistic method to distinguish oil spills from other similar oceanic features in marine (SAR) images has been developed and tested. The method uses statistical information obtained from previous measurements of physical and geometrical characteristics for both oil spill and natural features. A sample image is evaluated using a procedure to determine the probability that it is an oil spill. The classification-algorithm performance was evaluated using a test dataset of SAR images containing hundreds of examples of oil spills and of features exhibiting characteristics similar to oil spills (look-alike): more than 80% of the samples were classified correctly. The reliability of the method was then determined using a new dataset and similar results were obtained. The developed methodology and its capability in recognizing oil spills among look-alike are illustrated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
