Skylight applies Machine learning, and most often, “Deep Learning” to two types of data-- GPS and various remote sensing data (i.e. images).

Machine Learning is a helpful tool for analyzing these data because manual analysis of vessel tracks or large images is time intensive and tedious. 

While Machine Learning has its limitations and requires a human to confirm outputs from Machine Learning and/or conduct further analysis, it can save analysts hours of time and provide initial insights.

AIS

AIS is a form of GPS, or positional data. Skylight uses this data to identify certain behaviors of vessels. 

You can learn more about this in the Events/Trajectory Events section of the Resource Center.

Remote sensed data

Skylight is able to detect vessels from a variety of sources of satellite imagery and other signals.

You can learn more about this in the Events/Vessel Detection Events section of the Resource Center.

To learn more, check out this article by the Skylight team at Ai2: Satellite Imagery and AI: A New Era in Ocean Conservation, from Research to Deployment and Impact.