What is the challenge that is being addressed?

There are numerous stakeholders that are interested in maritime situational awareness, but most base their operations on self-reported ship AIS data. If ships are noncompliant, this can become a safety & security problem, or a precursor to illegal activity like smuggling or oil bunkering.

Through sensor fusion of proprietary data from ISR patrols and ship detections from multiple satellite earth observation sensors, governments, traders, ports and other stakeholders can know and react to suspicious activity at sea. Anchorage areas were chosen due to their fairly static nature, to have the best case solution with the spatial resolution of the sensors in this project. 

What is the AI solution the project has implemented?

We have developed AI pipelines and data processing tools for detecting ships in satellite imagery and discovering their RF-transceivers, combined with data fusion of AIS data and other proprietary onboard ISR equipment. This allows to flag suspicious vessels, and identify them with efficient guided patrols.

Who helped implement the AI solution?

This solution is implemented in the context of SR4C3, a winning project from the AI4Copernicus 1st Open Call. The consortium consists of VAKE, Shipdetection.eu, and NSAS, while the results of the project are available here and here.