Imaging the Planet for a Sustainable Future

Prof. Gilberto CAMARA
Secretariat Director for the Group on Earth Observations

with the special participation of:

Prof. Martin Vetterli, EPFL President
Prof. Meredith Schuman, UZH
Prof. Devis Tuia, EPFL

Satellite images are the most comprehensive source of data about our environment; they provide information for measuring deforestation, crop production, food security, urban footprints, water scarcity, and land degradation. In recent years, space agencies have adopted open distribution policies and as result experts now have access to repeated acquisitions over the same areas; the resulting time series improve our understanding of ecological patterns and processes.

The availability of big Earth observation data has led the experts in the field to focus on the technologies of cloud computing, data cubes, and machine learning. However, continuous monitoring of land dynamics using remote sensing data differs from applications such as spam filters, automatic translation, and object detection. This talk will focus on the challenges of using machine learning to analyse large satellite image time series and argue that long-term progress will depend on a new generation of methods that combine machine learning with functional ecosystem models.