Detection of Oil Spill Events at Sea Using Machine Learning

Oil spill detection performed by using Machine Learning (ML) involves building and training a model to identify and classify oil spill images from satellite or drone imagery. Synthetic Aperture Radar (SAR) is a technology that has been extensively used in oil spill detection. It is to detect and monitor tea-based oil pollution by airplanes and satellites furnished with SAR. It is necessary to see and spot oil spills in seas, oceans, rivers, and lakes. The ML model can be trained by images labeled with clean water and water with oil spills gathered from satellite and drone images with high resolution. In this proposed method, AIS data of ships were utilized to predict the oil spill events at sea before it happens. With the parameters like container details, type of oil, dispersion rate, skim rate, etc. The probability of an oil spill is detected by training a machine learning model. The software used for implementing the idea is Spyder IDE and algorithms like Linear Regression, Support Vector Machine (SVM), and Decision Tree were employed here. The major challenge in oil spill detection is accuracy and time complexity.