Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.777873
Title: Offshore oil spill detection using synthetic aperture radar
Author: Singha, Suman
ISNI:       0000 0004 7963 6414
Awarding Body: University of Hull
Current Institution: University of Hull
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Among the different types of marine pollution, oil spill has been considered as a major threat to the sea ecosystems. The source of the oil pollution can be located on the mainland or directly at sea. The sources of oil pollution at sea are discharges coming from ships, offshore platforms or natural seepage from sea bed. Oil pollution from sea-based sources can be accidental or deliberate. Different sensors to detect and monitor oil spills could be onboard vessels, aircraft, or satellites. Vessels equipped with specialised radars, can detect oil at sea but they can cover a very limited area. One of the established ways to monitor sea-based oil pollution is the use of satellites equipped with Synthetic Aperture Radar (SAR). The aim of the work presented in this thesis is to identify optimum set of feature extracted parameters and implement methods at various stages for oil spill detection from Synthetic Aperture Radar (SAR) imagery. More than 200 images of ERS-2, ENVSAT and RADARSAT 2 SAR sensor have been used to assess proposed feature vector for oil spill detection methodology, which involves three stages: segmentation for dark spot detection, feature extraction and classification of feature vector. Unfortunately oil spill is not only the phenomenon that can create a dark spot in SAR imagery. There are several others meteorological and oceanographic and wind induced phenomena which may lead to a dark spot in SAR imagery. Therefore, these dark objects also appear similar to the dark spot due to oil spill and are called as look-alikes. These look-alikes thus cause difficulty in detecting oil spill spots as their primary characteristic similar to oil spill spots. To get over this difficulty, feature extraction becomes important; a stage which may involve selection of appropriate feature extraction parameters. The main objective of this dissertation is to identify the optimum feature vector in order to segregate oil spill and 'look-alike' spots. A total of 44 Feature extracted parameters have been studied. For segmentation, four methods; based on edge detection, adaptive theresholding, artificial neural network (ANN) segmentation and the other on contrast split segmentation have been implemented. Spot features are extracted from both the dark spots themselves and their surroundings. Classification stage was performed using two different classification techniques, first one is based on ANN and the other based on a two-stage processing that combines classification tree analysis and fuzzy logic. A modified feature vector, including both new and improved features, is suggested for better description of different types of dark spots. An ANN classifier using full spectrum of feature parameters has also been developed and evaluated. The implemented methodology appears promising in detecting dark spots and discriminating oil spills from look-alikes and processing time is well below any operational service requirements.
Supervisor: Bellerby, T. J. Sponsor: University of Hull ; European Maritime Safety Agency ; European Commission Joint Research Centre
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.777873  DOI: Not available
Keywords: Geography
Share: