A Novel Sinergy Between Remote Sensing and GIS for Oil Spill Detection on Satellite Imagery

The large amount of images available today, thanks to the increasing of the number of orbiting EO satellites (Earth Observation Systems), which are able to provide information of every region of the Earth, represents an indispensable instrument for monitoring any terrestrial ecosystem. EO systems allow to detect and follow fast changing phenomena (like natural and anthropic disasters) providing the needed information for planning the necessary measures and reduce the impact of these events. This paper aims at showing the results obtained through the transformation of Mathematical Morphology algorithms (which have already shown their effectiveness) in IDL based algorithms (using ENVI EX rule sets) that can provide identical performances. The direct execution of these algorithms in a GIS environment, taking into account topology restrictions, allows to directly generate maps containing potential oil spills isolated on satellite images (exploiting the geometrical and physical characteristics of oil spills) together with contextual information on the same maps (typical of the GIS environment, as the wind regime, presence of vessels, etc) in order to assign a probability to candidate spots and create a much more accurate oil spill detection process.