Solar Image Restoration with the CycleGAN Based on Multi-fractal Properties of Texture Features

Jia, Peng; Huang, Yi; Cai, Bojun; Cai, Dongmei

Abstract

Texture is one of the most obvious characteristics in solar images and it is normally described by texture features. Because textures from solar images of the same wavelength are similar, we assume that texture features of solar images are multi-fractals. Based on this assumption, we propose a pure data-based image restoration method: with several high-resolution solar images as references, we use the Cycle-Consistent Adversarial Network to restore blurred images of the same steady physical process, in the same wavelength obtained by the same telescope. We test our method with simulated and real observation data and find that our method can improve the spatial resolution of solar images, without loss of any frames. Because our method does not need a paired training set or additional instruments, it can be used as a post-processing method for solar images obtained by either seeing-limited telescopes or telescopes with ground-layer adaptive optic systems.

Citation

Jia, P., Huang, Y., Cai, B., & Cai, D. (2019). Solar Image Restoration with the CycleGAN Based on Multi-fractal Properties of Texture Features. Astrophysical Journal Letters, 881(2), Article L30. https://doi.org/10.3847/2041-8213/ab365f

• Journal Article Type: Article
• Acceptance Date: Jul 26, 2019
• Online Publication Date: Aug 19, 2019
• Publication Date: Aug 19, 2019
• Deposit Date: Sep 4, 2019
• Publicly Available Date: Sep 4, 2019
• Journal: Astrophysical Journal Letters
• Print ISSN: 2041-8205
• Electronic ISSN: 2041-8213
• Publisher: American Astronomical Society
• Peer Reviewed: Peer Reviewed
• Volume: 881
• Issue: 2
• Article Number: L30
• DOI: https://doi.org/10.3847/2041-8213/ab365f
• Public URL: https://durham-repository.worktribe.com/output/1294086