Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes
NAGIOS: RODERIC FUNCIONANDO

Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes

DSpace Repository

Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes

Show simple item record

dc.contributor.author Mateo-Sanchis, Anna
dc.contributor.author Muñoz Marí, Jordi
dc.contributor.author Campos Taberner, Manuel
dc.contributor.author García Haro, Francisco Javier
dc.contributor.author Camps Valls, Gustavo
dc.date.accessioned 2020-12-12T10:48:19Z
dc.date.available 2020-12-13T05:45:07Z
dc.date.issued 2018 es_ES
dc.identifier.uri https://hdl.handle.net/10550/76651
dc.description.abstract 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. es_ES
dc.description.sponsorship CICYT TIN2015-64210-R es_ES
dc.language.iso en es_ES
dc.relation.ispartofseries 2018 IEEE International Geoscience and Remote Sensing Symposium;
dc.source A. Mateo-Sanchis, J. Muñoz-Marí, M. Campos-Taberner, J. García-Haro and G. Camps-Valls, "Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes," IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, 2018, pp. 4039-4042, doi: 10.1109/IGARSS.2018.8519254. es_ES
dc.subject Time series es_ES
dc.subject remote sensing es_ES
dc.subject machine learning es_ES
dc.subject gaussian processes es_ES
dc.title Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes es_ES
dc.type info:eu-repo/semantics/lecture es_ES
dc.subject.unesco UNESCO::CIENCIAS DE LA TIERRA Y DEL ESPACIO es_ES
dc.identifier.doi https://doi.org/10.1109/IGARSS.2018.8519254 es_ES
dc.description.abstractenglish 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. es_ES
dc.accrualMethod - es_ES
dc.embargo.terms 0 days es_ES

View       (258.6Kb)

This item appears in the following Collection(s)

Show simple item record

Search DSpace

Advanced Search

Browse

Statistics