UNSUPERVISED OIL SPILL DETECTION IN SAR IMAGERY THROUGH AN ESTIMATOR OF LOCAL REGULARITY

摘要

Since surfactive substances damp the short-scale waves,rnthe presence of oil in the sea surface is revealed as arndark patch in the SAR image. Hence, SAR systems arerneffectively suitable for oil spills identification on oceanrnsurfaces. In fact, SAR technology has been confirmed asrnone of the most appropriate for the detection of slicks.rnNevertheless, automatic segmentation in oceanic SARrnimagery is still troublesome and specific data miningrnmethods are still to be provided. In particular, in therncontext of pollution monitoring in the ocean, the mainrndrawback of computerised schemes is the discriminationrnof oil spills and look-alikes. This paper aims at sheddingrnlight on the possibility of solving the disambiguationrnissue through a multiscale estimation of the localrnregularity.