KRÖBER, F.; GARCIA, G. Fernandez; GUIOTTE, F.; DELERUE, Florian; CORPETTI, T.; LEFÈVRE, S.

doi:10.1109/IGARSS52108.2023.10281513

Voir/Ouvrir

IGARSS_Felix-1.pdf (4.213Mo)

Métadonnées

Afficher la notice complète

Licence d’utilisation du document

KRÖBER, F.
Littoral, Environnement, Télédétection, Géomatique [LETG - Rennes ]

GARCIA, G. Fernandez
Littoral, Environnement, Télédétection, Géomatique [LETG - Rennes ]

GUIOTTE, F.
L'Avion Jaune

Langue

Communication dans un congrès

Ce document a été publié dans

IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, 2023-07-16, Pasadena. p. 3322-3325

IEEE

Résumé en anglais

This study aims at recovering above-ground biomass information from ultra-high resolution UAV RGB-NIR orthophotos. We focus on a realistic scenario where a limited number of training samples for a landscape with heterogeneous herbaceous vegetation is given. Consequently, we explore different machine learning methods explicitly addressing the limitations of small training samples and compare their predictions quantitatively and qualitatively. Our results show that random forest models perform similarly well to deep learning models. While simpler machine learning models may, therefore, still be preferable, our study also points the way to promising architectures and regularisation techniques for deep learning approaches. Beyond vegetation cover, accurate regression of other variables, including vegetation height, volume and biomass remains a difficult task regardless of the model choice.< Réduire

Mots clés en anglais

vegetation biomass regression random forest deep learning semi-supervised learning transfer learning UAV

vegetation biomass regression

random forest

deep learning

semi-supervised learning

transfer learning

UAV

vegetation biomass regression random forest deep learning semi-supervised learning transfer learning UAV