Sim2Real SAR Image Restoration: Metadata-Driven Models for Joint Despeckling and Sidelobes Reduction
By: Antoine De Paepe , Pascal Nguyen , Michael Mabelle and more
Potential Business Impact:
Cleans up blurry satellite pictures for better views.
Synthetic aperture radar (SAR) provides valuable information about the Earth's surface under all weather and illumination conditions. However, the inherent phenomenon of speckle and the presence of sidelobes around bright targets pose challenges for accurate interpretation of SAR imagery. Most existing SAR image restoration methods address despeckling and sidelobes reduction as separate tasks. In this paper, we propose a unified framework that jointly performs both tasks using neural networks (NNs) trained on a realistic SAR simulated dataset generated with MOCEM. Inference can then be performed on real SAR images, demonstrating effective simulation to real (Sim2Real) transferability. Additionally, we incorporate acquisition metadata as auxiliary input to the NNs, demonstrating improved restoration performance.
Similar Papers
Towards Frequency-Adaptive Learning for SAR Despeckling
CV and Pattern Recognition
Cleans up blurry satellite pictures for better detail.
Lightweight CNNs for Embedded SAR Ship Target Detection and Classification
CV and Pattern Recognition
Lets satellites spot ships faster from space.
SARMAE: Masked Autoencoder for SAR Representation Learning
CV and Pattern Recognition
Makes satellites see clearly through clouds.