Image-Based Oil Spill Detection using Deep Learning Techniques: A Review

Oil spills represent a significant danger to marine ecosystems, leading to considerable environmental disasters that adversely affect biodiversity. These incidents, whether accidental or intentional, introduce harmful petroleum hydrocarbons into water bodies, endangering marine life. Various techniques have been developed for oil spill detection, including classification, zone segmentation, threshold segmentation, and edge detection. Recently, deep learning (DL) and machine learning (ML) have shown a lot of potential for successfully detecting and tracking oil spills. This research provides a valuable review of AI-based methodologies for oil spill detection from 2020 to 2024, emphasizing the use of satellite imagery, aerial photographs, and drone imagery for extensive monitoring of marine and coastal areas. The review also discusses the advantages and limitations of these approaches and suggests future directions to improve oil spill detection systems. Ultimately, this article serves as a valuable guideline for researchers providing innovative solutions to mitigate oil spill impacts, aiding in better environmental management and policy development.