Learning UAV-Based Above-Ground Biomass Regression Models in Sparse Training Data Environments
| dc.rights.license | open | en_US |
| hal.structure.identifier | Littoral, Environnement, Télédétection, Géomatique [LETG - Rennes ] | |
| dc.contributor.author | KRÖBER, F. | |
| hal.structure.identifier | Littoral, Environnement, Télédétection, Géomatique [LETG - Rennes ] | |
| dc.contributor.author | GARCIA, G. Fernandez | |
| hal.structure.identifier | L'Avion Jaune | |
| dc.contributor.author | GUIOTTE, F. | |
| hal.structure.identifier | Environnements et Paléoenvironnements OCéaniques [EPOC] | |
| dc.contributor.author | DELERUE, Florian | |
| hal.structure.identifier | Littoral, Environnement, Télédétection, Géomatique [LETG - Rennes ] | |
| dc.contributor.author | CORPETTI, T. | |
| hal.structure.identifier | Observation de l’environnement par imagerie complexe [OBELIX] | |
| dc.contributor.author | LEFÈVRE, S. | |
| dc.date.accessioned | 2024-03-07T09:07:55Z | |
| dc.date.available | 2024-03-07T09:07:55Z | |
| dc.date.conference | 2023-07-16 | |
| dc.identifier.uri | https://oskar-bordeaux.fr/handle/20.500.12278/188614 | |
| dc.description.abstractEn | 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. | |
| dc.language.iso | EN | en_US |
| dc.publisher | IEEE | en_US |
| dc.subject.en | vegetation biomass regression random forest deep learning semi-supervised learning transfer learning UAV | |
| dc.subject.en | vegetation biomass regression | |
| dc.subject.en | random forest | |
| dc.subject.en | deep learning | |
| dc.subject.en | semi-supervised learning | |
| dc.subject.en | transfer learning | |
| dc.subject.en | UAV | |
| dc.subject.en | vegetation biomass regression random forest deep learning semi-supervised learning transfer learning UAV | |
| dc.title.en | Learning UAV-Based Above-Ground Biomass Regression Models in Sparse Training Data Environments | |
| dc.type | Communication dans un congrès | en_US |
| dc.identifier.doi | 10.1109/IGARSS52108.2023.10281513 | en_US |
| dc.subject.hal | Sciences de l'environnement | en_US |
| bordeaux.page | 3322-3325 | en_US |
| bordeaux.hal.laboratories | EPOC : Environnements et Paléoenvironnements Océaniques et Continentaux - UMR 5805 | en_US |
| bordeaux.institution | Université de Bordeaux | en_US |
| bordeaux.institution | CNRS | en_US |
| bordeaux.conference.title | IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium | en_US |
| bordeaux.team | ECOBIOC | en_US |
| bordeaux.conference.city | Pasadena | en_US |
| bordeaux.import.source | hal | |
| hal.identifier | hal-04310706 | |
| hal.version | 1 | |
| hal.invited | non | en_US |
| hal.proceedings | non | en_US |
| hal.conference.end | 2023-07-21 | |
| hal.popular | non | en_US |
| hal.audience | Internationale | en_US |
| hal.export | false | |
| workflow.import.source | hal | |
| dc.rights.cc | Pas de Licence CC | en_US |
| bordeaux.COinS | ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.spage=3322-3325&rft.epage=3322-3325&rft.au=KR%C3%96BER,%20F.&GARCIA,%20G.%20Fernandez&GUIOTTE,%20F.&DELERUE,%20Florian&CORPETTI,%20T.&rft.genre=unknown |
