Detection of oil spills by airborne sensors

Abstract

Aerial surveillance of marine oil spills has become a multinational effort, especially for the deterrence of potential polluters and support of oil spill clean-up crews. For many years, the main effort has been directed towards developing sensors with enhanced spill monitoring capabilities. These sensors can be divided into different classes, namely basic sensors and advanced sensors including thermal imagers with active laser illumination. Recently, more attention has been paid to the automated derivation of high-level information from airborne remotely sensed multispectral oil spill data. In this paper, we present an algorithmic framework for automated analysis and fusion of multi-spectral oil spill data acquired by a near range sensor suite. It is shown how infrared/ultraviolet, microwave radiometer and imaging laser fluorosensor data can be converted into high-level information using single-sensor data processing and multi-sensor data fusion. It is also explained how this knowledge can, in combination with external information processing, improve the usability of the maritime surveillance system.