Oil Spill Detection based on Machine Learning and Deep Learning: A Review

For many years, oil spills have posed a huge and inescapable threat to the seas and oceans. Hence, it is noted that oil spills caused by the purposeful or unintentional discharge of liquid petroleum hydrocarbons into water are to blame for a number of ecological disasters, which disrupt the marine life cycle as well as degrade the productivity and quality of the marine environment, posing significant environmental dangers. There exist various methods, for example, zone segmentation, edge detection and threshold segmentation to address oil spill problem. Currently Machine Learning (ML) and Deep Learning (DL) are used as one of the efforts identify oil spills based on Artificial Intelligence (AI) because they are the most significant techniques and playing a significant role in the oil spills’ monitoring and accurate detection. This paper summary the classification of modern methods for detecting oil spills in the time period (2018-2021) especially explains the utilization of ML and DL methods to address the problem through the presentation and analysis. On the other hand, describe the advantage and disadvantage of these studies. In addition to denote the ideas for a future research direction to develop oil spill detection.