SAR despeckling is a problem of paramount importance in remote sensing, since it represents the first step of many scene analysis algorithms. Recently, deep learning techniques have outperformed classical model-based despeckling algorithms. However, such methods require clean ground truth images for training, thus resorting to synthetically speckled optical images since clean SAR images cannot be acquired. In this paper, inspired by recent works on blind-spot denoising networks, we propose a self-supervised Bayesian despeckling method. The proposed method is trained employing only noisy images and can therefore learn features of real SAR images rather than synthetic data. We show that the performance of the proposed network is very close to the supervised training approach on synthetic data and competitive on real data.

Towards Deep Unsupervised SAR Despeckling with Blind-Spot Convolutional Neural Networks / Molini, Andrea Bordone; Valsesia, Diego; Fracastoro, Giulia; Magli, Enrico. - ELETTRONICO. - (2020), pp. 2507-2510. ((Intervento presentato al convegno International Geoscience and Remote Sensing Symposium (IGARSS) [10.1109/IGARSS39084.2020.9324183].

Towards Deep Unsupervised SAR Despeckling with Blind-Spot Convolutional Neural Networks

Molini, Andrea Bordone;Valsesia, Diego;Fracastoro, Giulia;Magli Enrico
2020

Abstract

SAR despeckling is a problem of paramount importance in remote sensing, since it represents the first step of many scene analysis algorithms. Recently, deep learning techniques have outperformed classical model-based despeckling algorithms. However, such methods require clean ground truth images for training, thus resorting to synthetically speckled optical images since clean SAR images cannot be acquired. In this paper, inspired by recent works on blind-spot denoising networks, we propose a self-supervised Bayesian despeckling method. The proposed method is trained employing only noisy images and can therefore learn features of real SAR images rather than synthetic data. We show that the performance of the proposed network is very close to the supervised training approach on synthetic data and competitive on real data.
File in questo prodotto:
File Dimensione Formato  
SAR_Despeckling.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 1 MB
Formato Adobe PDF
1 MB Adobe PDF Visualizza/Apri
Valsesia-Towards.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 1.04 MB
Formato Adobe PDF
1.04 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11583/2844380