Image quality evaluation and fast masking with deep neural networks

Yu Song; Runyu Ning; Jiameng Lv; Peng Jia

doi:10.1117/12.2637005

29 August 2022 Image quality evaluation and fast masking with deep neural networks

Yu Song, Runyu Ning, Jiameng Lv, Peng Jia

Proceedings Volume 12189, Software and Cyberinfrastructure for Astronomy VII; 121890K (2022) https://doi.org/10.1117/12.2637005
Event: SPIE Astronomical Telescopes + Instrumentation, 2022, Montréal, Québec, Canada

Abstract

As more and more images are obtained by astronomical observations, a fast image quality evaluation algorithm is required for data processing pipelines. The image quality evaluation algorithm should be able to recognize blur or noise levels according to scientists’ requirements and further mask parts of images with low qualities. In this paper, we introduce a deep learning based image quality evaluation and fast masking algorithm. Our algorithm uses an auto-encoder neural network to obtain blur or noise levels and we further use blur or noise levels to generate mask maps for input images. Tested with simulated and real data, our algorithm could provide reliable results with small amount of images as the training set. Our algorithm could be used as a reliable image mask algorithm for different image processing pipelines.

Conference Presentation

Citation Download Citation

Yu Song, Runyu Ning, Jiameng Lv, and Peng Jia "Image quality evaluation and fast masking with deep neural networks", Proc. SPIE 12189, Software and Cyberinfrastructure for Astronomy VII, 121890K (29 August 2022); https://doi.org/10.1117/12.2637005

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