AstroSURE: Learning to Remove Noise from Astronomical Images Without Ground Truth Data
For astronomers, it assesses unsupervised denoising methods to enhance object detection without requiring clean training data, but results are incremental and domain-dependent.
The paper evaluates deep-learning denoising methods (Noise2Noise, SURE, blind-spot) that can be trained without clean ground-truth images for astronomical imaging, showing improved faint-source detectability on HST data but limited transfer to CFHT data.
In astronomical imaging, the low photon count of exposures necessitates extensive post-processing steps, including contamination removal and denoising. This paper evaluates deep-learning denoising methods that can be trained without clean ground-truth images and assesses their utility for detection11 oriented analysis of astronomical data. We adapt and compare Noise2Noise, Stein's Unbiased Risk Estimator, and blind-spot-based methods using synthetic data and real observations from the Hubble Space Telescope (HST) and the Canada-France-Hawaii Telescope (CFHT). Performance is evaluated using object-detection metrics, including correct detection rate and false alarm rate, together with image-based metrics and pixel-distribution diagnostics. The results show that these methods can improve faint-source detectability relative to the original noisy images, with encouraging gains on HST data after domain-consistent initialization, while transfer to CFHT data is more limited, highlighting the importance of instrument/domain similarity for unsupervised adaptation.