Rapport de recherche interne (Rapports)
Data Augmentation in Earth Observation: A Diffusion Model Approach
DE JESUS SOUSA, Tiago Alexandre; RIES, Benoit; GUELFI, Nicolas
2024
 

Documents


Texte intégral
main.pdf
Preprint Auteur (799.31 kB)
Demander un accès

Tous les documents dans ORBilu sont protégés par une licence d'utilisation.

Envoyer vers



Détails



Mots-clés :
Data Augmentation; Diffusion Model; Earth Observation; Remote Sensing; Satellite Imagery; Deep Learning
Résumé :
[en] The scarcity of high-quality Earth Observation (EO) imagery poses a significant challenge, despite its critical role in enabling precise analysis and informed decision-making across various sectors. This scarcity is primarily due to atmospheric conditions, seasonal variations, and limited geographical coverage, which complicates the application of Artificial Intelligence (AI) in EO. Data augmentation, a widely used technique in AI that involves generating additional data mainly through parameterized image transformations, has been employed to increase the volume and diversity of data. However, this method often falls short in generating sufficient diversity across key semantic axes, adversely affecting the accuracy of EO applications. To address this issue, we propose a novel four-stage approach aimed at improving the diversity of augmented data by integrating diffusion models. Our approach employs meta-prompts for instruction generation, harnesses general-purpose vision-language models for generating rich captions, fine-tunes an Earth Observation diffusion model, and iteratively augments data. We conducted extensive experiments using four different data augmentation techniques, and our approach consistently demonstrated improvements, outperforming the established augmentation methods, revealing its effectiveness in generating semantically rich and diverse EO images.
Disciplines :
Sciences informatiques
Auteur, co-auteur :
DE JESUS SOUSA, Tiago Alexandre  ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
RIES, Benoit ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
GUELFI, Nicolas ;  University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Computer Science (DCS)
Co-auteurs externes :
no
Langue du document :
Anglais
Titre :
Data Augmentation in Earth Observation: A Diffusion Model Approach
Date de publication/diffusion :
29 mars 2024
Nombre de pages :
15
Focus Area :
Sustainable Development
Objectif de développement durable (ODD) :
15. Vie terrestre
Disponible sur ORBilu :
depuis le 02 avril 2024

Statistiques


Nombre de vues
207 (dont 30 Unilu)
Nombre de téléchargements
4 (dont 4 Unilu)

Bibliographie


Publications similaires



Contacter ORBilu