Noise2Astro: Astronomical Image Denoising With Self-Supervised NeuralNetworks

TL;DR

This paper explores the use of self-supervised CNN algorithms, specifically Noise2Noise, for denoising astronomical images, demonstrating high accuracy in recovering flux under simulated noise conditions.

Contribution

It investigates the feasibility of self-supervised CNNs like Noise2Noise for astronomical image denoising, showing promising results on simulated data.

Findings

High flux recovery accuracy for Poisson noise (98.13%)

Effective Gaussian noise denoising with 96.45% flux recovery

Demonstrates potential for automated astronomical data processing

Abstract

In observational astronomy, noise obscures signals of interest. Large-scale astronomical surveys are growing in size and complexity, which will produce more data and increase the workload of data processing. Developing automated tools, such as convolutional neural networks (CNN), for denoising has become a promising area of research. We investigate the feasibility of CNN-based self-supervised learning algorithms (e.g., Noise2Noise) for denoising astronomical images. We experimented with Noise2Noise on simulated noisy astronomical data. We evaluate the results based on the accuracy of recovering flux and morphology. This algorithm can well recover the flux for Poisson noise ( $98.13${\raisebox{0.5ex}{\tiny$^{+0.77}_{-0.90} $}$\large\%$}) and for Gaussian noise when image data has a smooth signal profile ($96.45${\raisebox{0.5ex}{\tiny$^{+0.80}_{-0.96} $}$\large\%$}).