Atmospheric turbulence restoration improves planetary image clarity
ASTRA-SR: Atmospheric Seeing and Turbulence Restoration for Astronomical Image Super-Resolution
Computer Vision and Pattern Recognition
Summary
Images of planets taken from ground-based telescopes often look blurry and noisy because Earth's atmosphere distorts the light before it reaches the camera. The authors present ASTRA-SR, a new method that cleans and sharpens these images using synthetic training data that mimics real atmospheric effects. ASTRA-SR works in steps, first reducing noise, then restoring details at different scales, resulting in clearer, higher-resolution images. This method performs better than previous techniques in making planetary images clearer.
What this means in practice
- •For astronomical imaging teams: Improve ground-based planetary images by restoring clarity despite atmospheric blur and noise using ASTRA-SR.
- •For satellite image analysts: Enhance planetary surface details in satellite imagery by applying physics-based super-resolution trained on synthetic turbulence data.
Authors
Xining Ge, Ziteng Cui, Shuhong Liu
Abstract
Ground-based planetary imaging suffers from atmospheric turbulence, sensor noise, and limited sampling, making restoration a joint denoising, deblurring, and super-resolution problem. We present ASTRA-SR, a blind single-frame restoration framework trained on a physics-grounded synthetic dataset. High-dynamic-range spacecraft RAW observations serve as clean sources, and paired LR inputs are synthesized using measured layer-integrated turbulence strengths, propagated moving phase screens, exposure-averaged spatially varying PSFs, and sensor noise.ASTRA-SR first estimates a noise-suppressed but blur-retaining LR image, then restores spatial structure through multiscale processing and reconstructs HR detail with serial spatial-amplitude refinement. It yields a 0.49 dB foreground PSNR gain over the strongest baseline approaches.