Security and AI in the management of Earth Observation data
Dr. Alen Berta
Observing the earth with the help of satellite sensors is mostly a data-centric task that relies on processing large amounts of information. Machine Learning is quickly becoming a key tool in the Earth Observation (EO) domain, just as in many other domains that process large amounts of data. However, due to the nature of how EO data is used, the data itself and the algorithms that process it must be secured.In this talk we discuss how machine learning can be used to secure and guarantee the authenticity of (publicly available) EO data. Furthermore, we also discuss the different security considerations needed to protect machine learning algorithms that work with EO data.