- Appropriate Machine Learning methods for EO data. The research objectives about this topic are devoted to advance the exploitation of:
- i) Satellite Image Time Series (SITS) data ;
- ii) Multi-source Earth Observation (EO) data analysis.
- New Learning paradigms to deal with EO data:
- i) Going further with the complementary exploitation of multiple EO data sources ;
- ii) Dealing with ground truth paucity leveraging semi-supervised / weakly supervised machine learning settings ;
- iii) Linking EO data with non-EO data via foundation models
- iv) Advancing the state of the art of (spatial/temporal/radiometric) transfer learning for EO data.
- Interaction between Domain expert and Machine Learning model for EO data:
- i) Integrate apriori knowledge (expert or biophysical) in the learning process of the ML model;
- ii) Design learning models that explicitly allow to interpret the decision process under different dimensions (i.e. temporal, spatial, etc…);
- iii) Interpret/Explain model decision making connections with expert knowledge.
APPLICATION examples
- Food Security
- Forest monitoring
- Biodiversity mapping and monitoring
- Soil Moisture and Irrigation mapping