Active learning on large hyperspectral datasets: a preprocessing method - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Active learning on large hyperspectral datasets: a preprocessing method

L. Risser
B. Berthelot
  • Fonction : Auteur
  • PersonId : 1143809

Résumé

Machine learning algorithms demonstrated promising results for hyperspectral semantic segmentation. However, they strongly rely on the quality of training datasets. As far as the annotation of hyperspectral images is often expensive and time-consuming, only a few thousand pixels can be labeled. In this context, active learning algorithms select the most informative pixels to be labeled. In the machine learning community, recent active learning methods have overcome the performance of conventional algorithms but do not always scale to large remote sensing images. Therefore, we introduce in this paper a preprocessing method that allows the use of computationally intensive active learning algorithms without significant impacts on their effectiveness.
Fichier principal
Vignette du fichier
isprs-archives-XLIII-B3-2022-435-2022.pdf (18.82 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-03700279 , version 1 (21-06-2022)

Licence

Paternité

Identifiants

Citer

R. Thoreau, Véronique Achard, L. Risser, B. Berthelot, Xavier Briottet. Active learning on large hyperspectral datasets: a preprocessing method. XXIV ISPRS Congress, Jun 2022, Nice, France. pp.435-442, ⟨10.5194/isprs-archives-XLIII-B3-2022-435-2022⟩. ⟨hal-03700279⟩
43 Consultations
12 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More