Parsimonious Gaussian process models for the classification of hyperspectral remote sensing images

Mathieu Fauvel 1 Charles Bouveyron 2 Stephane Girard 3
1 DYNAFOR - Dynamiques Forestières dans l'Espace Rural
2 MAP5 - MAP5 - Mathématiques Appliquées à Paris 5
3 MISTIS - Modelling and Inference of Complex and Structured Stochastic Systems
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : A family of parsimonious Gaussian process models for classification is proposed in this letter. A subspace assumption is used to build these models in the kernel feature space. By constraining some parameters of the models to be common between classes, parsimony is controlled. Experimental results are given for three real hyperspectral data sets, and comparisons are done with three others classifiers. The proposed models show good results in terms of classification accuracy and processing time.
Type de document :
Article dans une revue
IEEE Geoscience and Remote Sensing Letters, IEEE - Institute of Electrical and Electronics Engineers, 2015, 12 (12), pp.2423-2427. <10.1109/LGRS.2015.2481321>
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01203269
Contributeur : Charles Bouveyron <>
Soumis le : jeudi 24 septembre 2015 - 21:39:48
Dernière modification le : vendredi 27 janvier 2017 - 18:04:28
Document(s) archivé(s) le : mardi 29 décembre 2015 - 09:21:14

Fichier

grsl_fauvel_pgpda.pdf
Fichiers produits par l'(les) auteur(s)

Licence

Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Collections

INRIA | INRA | INSMI | UNIV-PARIS5 | MAP5 | LJK | UGA | LJK_PS | LJK_PS_MISTIS | USPC

Citation

Mathieu Fauvel, Charles Bouveyron, Stephane Girard. Parsimonious Gaussian process models for the classification of hyperspectral remote sensing images. IEEE Geoscience and Remote Sensing Letters, IEEE - Institute of Electrical and Electronics Engineers, 2015, 12 (12), pp.2423-2427. <10.1109/LGRS.2015.2481321>. <hal-01203269>

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de
la notice

722

Téléchargements du document

193

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD