6533b884fe1ef96bd12e07ad
RESEARCH PRODUCT
Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes
Anna Mateo-sanchisJordi Muñoz MaríManuel Campos TabernerFrancisco Javier García HaroGustau Camps-vallssubject
remote sensingTime seriesmachine learninggaussian processes:CIENCIAS DE LA TIERRA Y DEL ESPACIO [UNESCO]UNESCO::CIENCIAS DE LA TIERRA Y DEL ESPACIOdescription
In this work we evaluate multi-output (MO) Gaussian Process (GP) models based on the linear model of coregionalization (LMC) for estimation of biophysical parameter variables under a gap filling setup. In particular, we focus on LAI and fAPAR over rice areas. We show how this problem cannot be solved with standard single-output (SO) GP models, and how the proposed MO-GP models are able to successfully predict these variables even in high missing data regimes, by implicitly performing an across-domain information transfer. CICYT TIN2015-64210-R In this work we evaluate multi-output (MO) Gaussian Process (GP) models based on the linear model of coregionalization (LMC) for estimation of biophysical parameter variables under a gap filling setup. In particular, we focus on LAI and fAPAR over rice areas. We show how this problem cannot be solved with standard single-output (SO) GP models, and how the proposed MO-GP models are able to successfully predict these variables even in high missing data regimes, by implicitly performing an across-domain information transfer.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2018-01-01 |