A general purpose image restoration method with deep neural network and active learning

Jiameng Lv; Runyu Ning; Yu Song; Peng Jia

doi:10.1117/12.2637111

29 August 2022 A general purpose image restoration method with deep neural network and active learning

Jiameng Lv, Runyu Ning, Yu Song, Peng Jia

Proceedings Volume 12189, Software and Cyberinfrastructure for Astronomy VII; 121891U (2022) https://doi.org/10.1117/12.2637111
Event: SPIE Astronomical Telescopes + Instrumentation, 2022, Montréal, Québec, Canada

Conference Poster

Abstract

The blurred range of astronomical image data we observe is usually uncertain, Due to the complex space environment, random noise, unpredictable atmospheric turbulence and other external factors. We usually use ground-based large aperture optical telescopes to observe astronomical images, which are mainly affected by atmospheric turbulence. Therefore, the restoration of astronomical images under the influence of arbitrary atmospheric turbulence is of great significance for the theoretical development and technological progress of astronomy. In this paper, a novel astronomical image restoration algorithm is proposed, which connects the deep learning based image restoration algorithm with the data generation method. The algorithm could effectively restore images within predefined blur or noise levels. We use long exposure galaxy images and short exposure Solar images to test the algorithm. We find that a well trained algorithm can restore these images.

Conference Presentation

Citation Download Citation

Jiameng Lv, Runyu Ning, Yu Song, and Peng Jia "A general purpose image restoration method with deep neural network and active learning", Proc. SPIE 12189, Software and Cyberinfrastructure for Astronomy VII, 121891U (29 August 2022); https://doi.org/10.1117/12.2637111

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available