AI4COPSEC will crucially rely on data sources availability for developing its six technological modular bricks.  The term “technological bricks” refers to modular, standalone technology components that each serve a specific function within the larger system architecture. These bricks are building blocks—designed to be developed, tested, and later integrated—to enhance the functionality and performance of existing Copernicus services or enable new ones, especially in the domain of border and maritime surveillance.

Figure: AI4COPSEC Data Sources Identification for its technological bricks

AI4COPSEC improves maritime monitoring by combining Copernicus data, AIS signals, and Melodi drifter measurements into one real-time system. Using Omni-Situ technology, it repurposes AIS to track surface currents and waves. The project also merges coastal and satellite AIS data and provides APIs to make this information easily accessible for faster, smarter maritime decisions.

AI4COPSEC enhances maritime surveillance by combining Copernicus satellite imagery with thermal data, using advanced AI models to detect ships and oil spills. Deep learning techniques such as CNNs and transformers analyse complex images, improving the accuracy and reliability of marine monitoring.

AI4COPSEC improves ship tracking by using satellite images (optical, SAR, thermal) together with AIS data and AI-based analysis. This makes it possible to follow ships even if they switch off their AIS. Machine learning models study movement patterns to help identify ships more accurately, supporting better maritime security.

AI4COPSEC spots maritime events like oil spills or distress calls in near real-time by scanning over 9,000 open sources, including news and public social media. HOZINT filters this data using keywords and AI tools, keeping only the most relevant content. Sources are rated for reliability, and a database of 7 million past alerts helps train the models. AIS data and language tools are used to pinpoint the time and place of each event.

AI4COPSEC spots unusual AIS signals by comparing ship movements with real-time sea conditions. It combines Copernicus data, drifter measurements, and AIS-derived currents and waves to train a machine learning model. This helps detect when a ship’s behaviour doesn’t match ocean conditions, flagging possible risks. The system improves monitoring and makes anomaly detection at sea more accurate and reliable.

AI4COPSEC improves drift prediction using two methods. One combines AIS and Copernicus data with high-resolution SST and ocean colour imagery via 4DVarNet, making current models more accurate for rescue and pollution response. The other uses a machine learning model trained on Melodi drifter data to better capture ocean behaviour over time. Melodi IoT drifters send real-time data on currents, waves, and temperature, helping to validate and improve predictions—boosting maritime awareness and response.