We are happy to announce the project kick-off of the ESA funded project RepreSent on 6.4.2022. The project runs from 1.4.2022 to 31.3.2023. This project is led by our Department “EO Data Science” at DLR, in collaboration with our partners from EPFL, VTT and e-geos. The main objective is to capitalize on the potential of AI and Earth observation (EO) by exploiting the non-supervised learning paradigms. To do so, RepreSent will bring forward non-supervised learning-based solutions for impactful use cases that use unlabeled EO data.
In particular the use cases will be
- Forest disturbance mapping
- Automated land cover mapping
- Anomaly detection in long time series
- Cloud detection and removal
- Forest biomass estimation
We are looking forward to a fruitful collaboration and to exciting new results in field of unsupervised learning in Earth Observation.