A New Object-oriented Methodology to Detect Oil Spills using ENVISAT Images

Several ASAR images from ENVISAT were tested for oil spill detection using a new object oriented approach. A new automated methodology for oil spill detection were previously introduced, by which full SAR high resolution image scenes can be processed. In the present paper the method is tested using full high resolution ENVISAT data. The methodology relies on the object oriented approach and profits of image segmentation techniques in order for dark formations to be detected. The detection of dark formations is based on a threshold definition which is fully adaptive to local contrast and brightness of large image segments. For the detection process, two empirical formulas were developed, which also permit the classification of oil spills according to their brightness. A fuzzy classification method is used to classify dark formations to oils spill or look-alikes. Dark formations are not isolated and features of both dark areas and sea environment are considered. Various sea environments which affect oil spill shape and boundaries are grouped in two knowledge bases, used for the classification of dark formations. The method’s accuracy was tested for ENVISAT images. Previously test for 12 ERS images saw more than 99% for oil spill accuracy, and close to 99% for look-alike accuracy. Fresh oil spills, fresh spills affected by natural phenomena, oil spills without clear stripping, small linear oil spills, oil spills with broken parts and amorphous oil spills can be successfully detected.

TOPOUZELIS Konstantinos;  KARATHANASSI Vassilia;  PAVLAKIS Petros;  ROKOS Demetrius; 

2007-10-15

ESA Communication Production Office

JRC37003

http://envisat.esa.int/envisatsymposium/proceedings/contents.html,    https://publications.jrc.ec.europa.eu/repository/handle/JRC37003,