Machine Learning Technologies for Enabling Earth Observation

Machine learning plays a vital role in the field of Earth observation. This role is sometimes a direct application of an existing machine learning methods, but more often than not, a novel approach is required owing to the individual nature of the application and the data. My research explores and develops machine learning technologies to automate the production of land cover maps of the Earth’s surface. Using satellite imagery as data, the algorithms developed facilitate the production of maps at unprecedented accuracy and scale.