Transformer-Based SAR Image Despeckling
TL
TL;DR
The proposed despeckling network comprises of a transformer-based encoder which allows the network to learn global dependencies between different image regions - aiding in better despekling.
Abstract
Synthetic Aperture Radar (SAR) images are usually degraded by a multiplicative noise known as speckle which makes processing and interpretation of SAR images difficult. In this paper, we introduce a transformer-based network for SAR image despeckling. The proposed despeckling network comprises of a transformer-based encoder which allows the network to learn global dependencies between different image regions - aiding in better despeckling. The network is trained end-to-end with synthetically generated speckled images using a composite loss function. Experiments show that the proposed method achieves significant improvements over traditional and convolutional neural network-based despeckling methods on both synthetic and real SAR images. Our code is available at: https://github.com/malshaV/sar_transformer
References16 items
1
MRDDANet: A Multiscale Residual Dense Dual Attention Network for SAR Image Denoising
2
SwinIR: Image Restoration Using Swin Transformer
3
Uformer: A General U-Shaped Transformer for Image Restoration
4
SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
5