In this paper we investigate the use of discriminative model learning through Convolutional Neural Networks (CNNs) for SAR image despeckling. The network uses a residual learning strategy, hence it does not recover the filtered image, but the speckle component, which is then subtracted from the noisy one. Training is carried out by considering a large multitemporal SAR image and its multilook version, in order to approximate a clean image. Experimental results, both on synthetic and real SAR data, show the method to achieve better performance with respect to state-of-the-art techniques.

SAR image despeckling through convolutional neural networks / Chierchia, Giovanni; Cozzolino, Davide; Poggi, Giovanni; Verdoliva, Luisa. - (2017). [10.1109/IGARSS.2017.8128234]

SAR image despeckling through convolutional neural networks

Davide Cozzolino;Giovanni Poggi;Luisa Verdoliva
2017

Abstract

In this paper we investigate the use of discriminative model learning through Convolutional Neural Networks (CNNs) for SAR image despeckling. The network uses a residual learning strategy, hence it does not recover the filtered image, but the speckle component, which is then subtracted from the noisy one. Training is carried out by considering a large multitemporal SAR image and its multilook version, in order to approximate a clean image. Experimental results, both on synthetic and real SAR data, show the method to achieve better performance with respect to state-of-the-art techniques.
2017
978-1-5090-4951-6
SAR image despeckling through convolutional neural networks / Chierchia, Giovanni; Cozzolino, Davide; Poggi, Giovanni; Verdoliva, Luisa. - (2017). [10.1109/IGARSS.2017.8128234]
File in questo prodotto:
Non ci sono file associati a questo prodotto.

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: https://hdl.handle.net/11588/703221
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 250
  • ???jsp.display-item.citation.isi??? 199
social impact