Oil Spill Detection using Satellite Images and AIS Data

oil spills are some of the most devastating marine environmental hazardous, their impacts span far and wide and cause serious damages to marine ecosystems, public health, and economy. Their rapid and accurate detection is crucial in order to provide timely response and containment. The current project proposes a smart system, which uses data from Automatic identification system (AIS), geo-referenced, and integrated with satellite imagery in order to detect oil spills and distress vessels. Through real-time vessel tracking and remote sensing, the system is expected to detect anomalies such as unusual movement of vessels or change in locations that may imply cases of distress or pollution. Using machines’ learning models like Support Vector Machines (SVM), Random Forest, as well as XGBoost, the model classifies patches of satellite images to detect the presence of oil spills. The backend is implemented using the Python language, and it offers a simple web interface for the real-time visualization and monitoring. This composite solution allows for environmental monitoring to make it easy to respond to oil-related incidents promptly and have sustainable maritime undertakings.