SAR Despeckling Using Overcomplete Convolutional Networks

2022/05/31 by Perera, Malsha V., Bandara, Wele Gedara Chaminda, Valanarasu, Jeya Maria Jose +1
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2205.15906

Abstract

Synthetic Aperture Radar (SAR) despeckling is an important problem in remote sensing as speckle degrades SAR images, affecting downstream tasks like detection and segmentation. Recent studies show that convolutional neural networks(CNNs) outperform classical despeckling methods. Traditional CNNs try to increase the receptive field size as the network goes deeper, thus extracting global features. However,speckle is relatively small, and increasing receptive field does not help in extracting speckle features. This study employs an overcomplete CNN architecture to focus on learning low-level features by restricting the receptive field. The proposed network consists of an overcomplete branch to focus on the local structures and an undercomplete branch that focuses on the global structures. We show that the proposed network improves despeckling performance compared to recent despeckling methods on synthetic and real SAR images.

Related

• SAR image despeckling through convolutional neural networks
• Speckle2Void: Deep Self-Supervised SAR Despeckling with Blind-Spot Convolutional Neural Networks
• Learning a Dilated Residual Network for SAR Image Despeckling
• Complexity Analysis of an Edge Preserving CNN SAR Despeckling Algorithm
• Blind SAR Image Despeckling Using Self-Supervised Dense Dilated Convolutional Neural Network