Title :
Algorithms for oil spill detection in Radarsat and ENVISAT SAR images
Author :
Schistad Solberg, A. ; Brekke, Camilla ; Solberg, Rune ; Husøy, Per Ove
Author_Institution :
Dept. of Inf., Oslo Univ.
Abstract :
We present algorithms for detection of oil spills in Radarsat and ENVISAT SAR images. The algorithms are trained on 60-100 images from each sensor from European water (mainly the North Sea and the Baltic Sea). Results show that they perform very well when applied to benchmark data sets, but in some cases additional information is needed to discriminate between oil slicks and algae, particularly in the Baltic Sea
Keywords :
benchmark testing; feature extraction; geophysical signal processing; image classification; image segmentation; oceanographic regions; oceanographic techniques; oils; radar detection; radar imaging; remote sensing by radar; seawater; synthetic aperture radar; Baltic Sea; ENVISAT SAR images; European water; North Sea; Radarsat images; algae; benchmark data sets; oil slicks; oil spill detection; Accidents; Aircraft; Algae; Backscatter; Image sensors; Inspection; Petroleum; Radar detection; Satellites; Speckle;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-8742-2
DOI :
10.1109/IGARSS.2004.1370264
