Learning Speckle Suppression in Sar Images Without Ground Truth: Application to Sentinel-1 Time-Series - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

Learning Speckle Suppression in Sar Images Without Ground Truth: Application to Sentinel-1 Time-Series

Abstract

This paper proposes a method of denoising SAR images, using a deep learning method, which takes advantage of the abundance of data to learn on large stacks of images of the same scene. The approach is based on the use of convolu-tional networks, used as auto-encoders. Learning is led on a large pile of images acquired on the same area, and assumes that the images of this stack differ only by the speckle noise. Several pairs of images are chosen randomly in the stack, and the network tries to predict the slave image from the master image. In this prediction, the network can not predict the noise because of its random nature. Also the application of this network to a new image fulfills the speckle filtering function. Results are given on Sentinel 1 images. They show that this approach is qualitatively competitive with literature.
Fichier principal
Vignette du fichier
DTIS19191.1570094372_preprint.pdf (3.68 Mo) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02329355 , version 1 (23-10-2019)

Identifiers

Cite

Alexandre Boulch, Pauline Trouvé, Elise Koeniguer, Fabrice Janez, Bertrand Le Saux. Learning Speckle Suppression in Sar Images Without Ground Truth: Application to Sentinel-1 Time-Series. 38th Annual IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2018, Jul 2018, VALENCIA, Spain. pp.2366-2369, ⟨10.1109/IGARSS.2018.8519370⟩. ⟨hal-02329355⟩
57 View
165 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More