Non-uniformity correction algorithm based on improved neural network

Xin Zhang; Huan Li; Juntao Hou; Dong Zhao; Huixin Zhou; Jiajia Zhang; Zhe Zhang; Kuanhong Cheng

doi:10.1117/12.2586410

12 March 2021 Non-uniformity correction algorithm based on improved neural network

Xin Zhang, Huan Li, Juntao Hou, Dong Zhao, Huixin Zhou, Jiajia Zhang, Zhe Zhang, Kuanhong Cheng

Proceedings Volume 11763, Seventh Symposium on Novel Photoelectronic Detection Technology and Applications; 117632P (2021) https://doi.org/10.1117/12.2586410
Event: Seventh Symposium on Novel Photoelectronic Detection Technology and Application 2020, 2020, Kunming, China

Abstract

Aiming at the problem of the traditional neural network for non-uniformity correction easy to cause ghosting artifacts and image blurring, an improved non-uniformity correction algorithm based on neural network is proposed. Firstly, a new fast trilateral filter is designed, which can be regarded as an edge-preserving smoothing operator. Secondly, in order to stabilize and accelerate the learning process, it adopts the self-adaptive learning rate and applies additional momentum factor to the neural network. Thirdly, in order to update the calibration parameters accurately, the local motion of different areas is judged carefully. The simulating experiments indicate that the proposed algorithm can suppress the ghosting artifacts and the image degradation. And it has better performance compared with other algorithms.

Citation Download Citation

Xin Zhang, Huan Li, Juntao Hou, Dong Zhao, Huixin Zhou, Jiajia Zhang, Zhe Zhang, and Kuanhong Cheng "Non-uniformity correction algorithm based on improved neural network", Proc. SPIE 11763, Seventh Symposium on Novel Photoelectronic Detection Technology and Applications, 117632P (12 March 2021); https://doi.org/10.1117/12.2586410

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