English

SAR image despeckling through convolutional neural networks

Computer Vision and Pattern Recognition 2017-05-04 v2

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.

Keywords

Cite

@article{arxiv.1704.00275,
  title  = {SAR image despeckling through convolutional neural networks},
  author = {G. Chierchia and D. Cozzolino and G. Poggi and L. Verdoliva},
  journal= {arXiv preprint arXiv:1704.00275},
  year   = {2017}
}

Comments

Accepted at 2017 IEEE International Geoscience and Remote Sensing Symposium, Fort Worth, Texas, July 23-28, 2017

R2 v1 2026-06-22T19:04:48.476Z