Non-parametric functional methods for hyperspectral image classification
Anthony Zullo
(1)
,
Mathieu Fauvel
(1)
,
Frédéric Ferraty
(2)
,
Goulard Michel
(1)
,
Philippe Vieu
(2)
Mathieu Fauvel
- Fonction : Auteur
- PersonId : 747895
- IdHAL : mathieu-fauvel
- ORCID : 0000-0002-3304-6932
- IdRef : 129264245
Frédéric Ferraty
- Fonction : Auteur
- PersonId : 15283
- IdHAL : frederic-ferraty
- ORCID : 0000-0002-5537-2667
- IdRef : 033150737
Goulard Michel
- Fonction : Auteur
- PersonId : 1147670
- IdHAL : michel-goulard-dyn
- ORCID : 0000-0001-7494-263X
Résumé
The objective of this article is to assess the relevance of a statistical method for hyperspectral image classification. We focus on the implementation of a functional method whose main objective is to consider each hyperspectrum as a continuous curve in order to predict its associated class. The implemented functional nonparametric discrimination method is a recently developed technique whose performance are greatly dependent on the choice of a "proximity measure". Behavior in practice of this method has been compared with three more standard others on two sets of hyperspectral data with supervised classification for 50 independent sets using a classification error rate criterion. Experimental results show that this method provides an interesting alternative to conventional methods.
Format du dépôt | Notice |
---|---|
Type de dépôt | Communication dans un congrès |
Titre |
en
Non-parametric functional methods for hyperspectral image classification
|
Résumé |
en
The objective of this article is to assess the relevance of a statistical method for hyperspectral image classification. We focus on the implementation of a functional method whose main objective is to consider each hyperspectrum as a continuous curve in order to predict its associated class. The implemented functional nonparametric discrimination method is a recently developed technique whose performance are greatly dependent on the choice of a "proximity measure". Behavior in practice of this method has been compared with three more standard others on two sets of hyperspectral data with supervised classification for 50 independent sets using a classification error rate criterion. Experimental results show that this method provides an interesting alternative to conventional methods.
|
Auteur(s) |
Anthony Zullo
1
, Mathieu Fauvel
1
, Frédéric Ferraty
2
, Goulard Michel
1
, Philippe Vieu
2
1
DYNAFOR -
Dynamiques Forestières dans l'Espace Rural
( 37920 )
- 24 Chemin de Borde-Rouge Auzeville CS 52627
31326 Castanet Tolosan Cedex
- France
2
IMT -
Institut de Mathématiques de Toulouse UMR5219
( 1954 )
- UPS IMT, F-31062 Toulouse Cedex 9,
INSA Toulouse, F-31077 Toulouse,
France UT1, F-31042 Toulouse,
France UT2, F-31058 Toulouse,
Téléphone : 05.61.55.67.90
- France
|
Audience |
Internationale
|
ISBN |
978-1-4799-5775-0
|
Date de publication |
2014
|
Ville |
Quebec
|
Pays |
Canada
|
Public visé |
Scientifique
|
Vulgarisation |
Non
|
ISSN |
2153-6996
|
Titre du congrès |
IEEE International Geoscience and Remote Sensing Symposium (IGARSS)
|
Date début congrès |
2014-07-13
|
Date fin congrès |
2014-07-18
|
Comité de lecture |
Oui
|
Page/Identifiant |
4
|
Invité |
Non
|
Source |
IEEE International Geoscience and Remote Sensing Symposium Proceedings
|
Lieu de publication |
NEW YORK (united states)
|
Langue du document |
Anglais
|
Actes |
Non
|
Domaine(s) |
|
Éditeur commercial |
|
Mots-clés |
en
Curse of dimensionality, hyperspectral image classification, nonparametric functional model, statistical method
|
DOI | 10.1109/IGARSS.2014.6947217 |
ProdINRA | 372145 |
UT key WOS | 000349688104179 |
Loading...