Semantic Segmentation of Sar Images Through Fully Convolutional Networks and Hierarchical Probabilistic Graphical Models - IDEX UCA JEDI Université Côte d'Azur Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Semantic Segmentation of Sar Images Through Fully Convolutional Networks and Hierarchical Probabilistic Graphical Models

Résumé

This paper addresses the semantic segmentation of synthetic aperture radar (SAR) images through the combination of fully convolutional networks (FCNs), hierarchical probabilistic graphical models (PGMs), and decision tree ensembles. The idea is to incorporate long-range spatial information together with the multiresolution information extracted by FCNs, through the multiresolution graph topology on which hierarchical PGMs can be efficiently formulated. The objective is to obtain accurate classification results with small datasets and reduce problems of spatial inconsistency. The experimental validation is conducted with several COSMO-SkyMed satellite images over Northern Italy. The results are significant, as the proposed method obtains more accurate classification results than the standard FCNs considered.
Fichier principal
Vignette du fichier
Martina_IGARSS2022.pdf (944.19 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03655029 , version 1 (29-04-2022)

Identifiants

Citer

Martina Pastorino, Gabriele Moser, Sebastiano Serpico, Josiane Zerubia. Semantic Segmentation of Sar Images Through Fully Convolutional Networks and Hierarchical Probabilistic Graphical Models. IEEE IGARSS 2022- International Geoscience and Remote Sensing Symposium, Jul 2022, Kuala Lumpur / Virtual, Malaysia. ⟨10.1109/IGARSS46834.2022.9883111⟩. ⟨hal-03655029⟩
133 Consultations
91 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More