Physics-constrained deep learning for biophysical parameter retrieval from Sentinel-2 images: inversion of the PROSAIL model - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail (Preprint/Prepublication) Année : 2023

Physics-constrained deep learning for biophysical parameter retrieval from Sentinel-2 images: inversion of the PROSAIL model

Yoël Zérah
  • Fonction : Auteur
  • PersonId : 1181481
  • IdHAL : yoel-zerah
Silvia Valero
  • Fonction : Auteur
  • PersonId : 937533
Jordi Inglada

Résumé

In this era of global warming, the regular and accurate mapping of vegetation conditions is essential for monitoring ecosystems, climate sustainability and biodiversity. In this context, this work proposes a physics-guided data-driven methodology to invert Radiative Transfer Models for the retrieval of vegetation biophysical variables. A hybrid paradigm is proposed by incorporating the physical model to be inverted into the design of a neural network architecture, which is trained by exploiting unlabeled satellite images. In this study, we show how the proposed strategy allows the simultaneous probabilistic inversion of all input PROSAIL model parameters by exploiting Sentinel-2 images. The interest of the proposed self-supervised learning strategy is corroborated by showing the limitations of existing simulation-trained machine learning algorithms. Results are assessed on LAI and CCC in-situ measurements collected on 4 different field campaigns over three European tests sites. Prediction accuracies are compared with performances reached by the well-established Simplified Level 2 Product Prototype Processor (SL2P). Obtained overall accuracies corroborate that the proposed methodology achieves performances equivalent to or better than the state-of-the-art methods.
Fichier principal
Vignette du fichier
manuscript.pdf (6.29 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Licence : CC BY - Paternité

Dates et versions

hal-04345612 , version 1 (14-12-2023)

Licence

Paternité - Pas d'utilisation commerciale - Partage selon les Conditions Initiales

Identifiants

  • HAL Id : hal-04345612 , version 1

Citer

Yoël Zérah, Silvia Valero, Jordi Inglada. Physics-constrained deep learning for biophysical parameter retrieval from Sentinel-2 images: inversion of the PROSAIL model. 2023. ⟨hal-04345612⟩
86 Consultations
43 Téléchargements

Partager

Gmail Facebook X LinkedIn More