An In-Depth Study of Oil Spill Detection Using Various Machine Learning Techniques

Abstract

Oil spills represent a significant threat to marine environments, causing ecological disasters and disrupting marine life cycles. This paper explores the environmental impact of oil spills, tracing their origins to the 1960s with the rise of supertankers. Despite stringent regulations, an annual transport of one million metric tons of oil into seas highlights the ongoing risk. To address this, the study proposes a structured approach for oil spill detection, encompassing data acquisition, preprocessing, feature extraction, and machine learning classification. The analysis emphasizes the diminishing occurrence of significant oil slicks, attributing this trend partly to regulatory efforts. The study highlights the growing role of deep learning in precise oil spill detection and calls for future research to address challenges, explore simplified detection techniques, and compare computational performance with traditional methods. This study summarizes the classification of modern ways for recognizing oil spills, as well as how machine learning techniques are applied to solve the problem through presentation and analysis. Conversely, discuss the benefits and drawbacks of these researches. In addition to indicating potential avenues for further investigation to advance the detection of oil spills.