Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5754642 | Remote Sensing Applications: Society and Environment | 2017 | 38 Pages |
Abstract
Offshore oil seeps occur in waters across the globe but locating them using in-situ methods requires a significant amount of time and cost. This motivated the design and implementation of an automatic method that can provide the location of offshore oil seeps from Synthetic Aperture Radar (SAR) data and resulted in the creation of the Automatic oil Seep Location Estimator (ASLE). Designed and implemented in the open source programming language Python, it uses a rule-based decision classifier to automatically discriminate between offshore oil slicks and other features visible in ocean SAR images. This paper reports how the detection and seep location estimation results improve when contextual wind information is added during object classification in the ASLE. The optimisation is discussed by comparing the new set of results to an already validated dataset of ENVISAT images of the Black Sea. Additionally, the paper reports the results of testing the ASLE on a dataset of RADARSAT images and the offshore oil seeps estimated in the Gulf of Mexico. The efficiency of ASLE with respect to other existing algorithms is also discussed.
Related Topics
Physical Sciences and Engineering
Earth and Planetary Sciences
Computers in Earth Sciences
Authors
Gopika Suresh, Christian Melsheimer, Ian R. MacDonald, Justus Notholt, Gerhard Bohrmann,