Advancing Surveillance in Maritime Domain with Ship Wake Detection

Le résumé fourni par la source

The detection of "dark" ships—vessels operating without Automatic Identification System (AIS) transponders—is crucial for maritime security and law enforcement. Traditional methods relying on optical and Synthetic Aperture Radar (SAR) imagery face limitations, particularly for small, fast-moving vessels. The UEIKAP project, funded under the Research Projects of National Interest (PRIN) by Italian Ministry of University and Research, is developing a novel approach by focusing on ship wakes, which provide valuable information about vessel characteristics and activity. This manuscript presents a comprehensive overview of the UEIKAP project’s approach, which leverages advanced deep learning techniques (DL) and data fusion from diverse sources, including optical and SAR satellite imagery, AIS data, meteo-marine measurements, and synthetic wake models, to develop a robust ship wake detection system. The system architecture is presented, featuring specialized modules which exploit the capabilities of Convolutional Neural Networks (CNNs). Additionally, the integration of contextual environmental data is detailed, demonstrating its role in enhancing detection accuracy under varying meteo-marine conditions. Experimental results from the Venice campaign in July 2024 are presented, demonstrating the system’s ability to detect small vessel wakes in optical imagery while highlighting the dependence of SARbased detection on wind conditions. These findings emphasize the importance of environmental factors and provide critical insights for refining detection algorithms. The UEIKAP project, scheduled to conclude in late 2025, plans to show a new approach for maritime domain awareness, showing the value of ship wake detection for monitoring vessel activity and enforcing international maritime laws.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.