In the next two weeks we'll have three FFT talks at the IPL.

You're invited to join!

9-Feb

11.30    "Spectral analysis of the short-term absorbance leaf mechanisms during quick photoprotection"
              Shari Van Wittenberghe  (*she'll bring belgium chocolate*)

I would like to present some spectral analysis which was focused on signal interpretation in terms of photochemistry up to now, with the defining of some spectral endmembers as you would call them. However, I think the data has more potential for further unmixing analysis!

12.30    "Physics-guided Machine Learning in New Orleans"
              Gustau Camps-Valls (*he'll bring valencian cacaus i tramussos*)

In this talk I'll briefly review what I learned in the AGU conference about combining machine learning and physical models for geosciences and Earth observation. This was actually done by attending talks and interacting with a group of people (NASA, Stanford, Minnessota and MIT) working in different related activities to my ERC grant. I'll review their current ideas, and point out the interesting relation to our methods: Joint Gaussian processes and Fair Kernel Learning.

12-Feb 

12:00    "Atmospheric radiative transfer models for dummies - Theory, practice and applications in EO"
              Jorge Vicent Servera (*he'll bring french wine*)

Atmospheric Radiative Transfer (RT) is one of the key elements to understand the interaction of light through the atmosphere with applications in Earth Observation such as instrument/mission design, atmospheric chemistry and atmospheric correction. However, the resolution of the RT equation is not straight-forward and most times complex to use in practice. A large number of atmospheric RT models (e.g. MODTRAN, 6SV...) have thus been developed to facilitate the simulation these radiative processes but their operation is still complex and required tailored modifications for some remote sensing applications (e.g. generation of large Look-Up Tables, model cross-comparison). The purpose of this talk is to present a newly developed software tool (ALG) that facilitates the user interaction with various atmospheric RT models. The talk will therefore give an overview of the atmospheric RT, models and give a tutorial of the ALG tool.