Empirical Analysis of Oil Spill Detection Methods
Date Issued
January 1, 2024
Abstract
Oil spills are a major source of marine pollution affecting the environment, economy, and marine ecosystems. Toxic chemicals from oil spills can remain in the ocean for years and even sink to the seabed, affecting sedimentation rates. Although many oil spills are caused by accident, some are caused intentionally by cargo ships dumping waste oil and bilge water. It is very difficult to locate, detect and remove oil from the ocean surface. However, regular monitoring can help prevent illegal dumping and aid remediation efforts. This work aims to detect oil spills in the North-Eastern part of Cyprus using a deep learning model. The results are compared with a conventional Adaptive Thresholding Algorithm. The comparisons demonstrate that the deep learning model has higher accuracy than the adaptive thresholding algorithm.
Funding
The present work was carried out in the framework of AI-OBSERVER project (https://ai-observer.eu/) titled "Enhancing Earth Observation capabilities of the Eratosthenes Centre of Excellence on Disaster Risk Reduction through Artificial Intelligence", that has received funding from the European Union’s Horizon Europe Framework Programme HORIZON-WIDERA-2021-ACCESS-03 (Twinning) under the Grant Agreement No 101079468. The authors also acknowledge the ‘EXCELSIOR’: ERATOSTHENES: Excellence Research Centre for Earth Surveillance and Space-Based Monitoring of the Environment H2020 Widespread Teaming project (www.excelsior2020.eu) in which the Eratosthenes CoE has been established. The ‘EXCELSIOR’ project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No 857510, from the Government of the Republic of Cyprus through the Directorate General for the European Programmes, Coordination and Development and the Cyprus University of Technology.