Data--driven Image Restoration with Option--driven Learning for Big and Small Astronomical Image Datasets

TL;DR

This paper introduces a novel GAN-based image restoration method for astronomical images that adapts to varying reference image availability, improving stability in diverse observational conditions.

Contribution

It proposes an option-driven learning approach that enhances GAN-based restoration for both big and small datasets, addressing the scarcity of paired training data.

Findings

Achieves stable restoration results across different reference image quantities.

Effective in real sky survey conditions with variable observation parameters.

Outperforms traditional supervised methods in unpaired data scenarios.

Abstract

Image restoration methods are commonly used to improve the quality of astronomical images. In recent years, developments of deep neural networks and increments of the number of astronomical images have evoked a lot of data--driven image restoration methods. However, most of these methods belong to supervised learning algorithms, which require paired images either from real observations or simulated data as training set. For some applications, it is hard to get enough paired images from real observations and simulated images are quite different from real observed ones. In this paper, we propose a new data--driven image restoration method based on generative adversarial networks with option--driven learning. Our method uses several high resolution images as references and applies different learning strategies when the number of reference images is different. For sky surveys with variable observation conditions, our method can obtain very stable image restoration results, regardless of the number of reference images.