Detail

Learning Spatio-Temporal Oceanographic Patterns

The main objective of this project is the development of Machine Learning methods for the automatic identification, analysis and prediction of oceanographic mesoscale phenomena in the Iberian Coastal Ocean (eddies, upwelling and coastal counter-current), including their spatio-temporal evolution, from remote sensing images. Dynamic versions of fuzzy clutstering will be investigated in the identification and tracking of these phenomena, and knowledge-aware neural networks will be studied for their classification and prediction incorporating domain knowledge.