Neural network non-uniformity correction for eliminating low frequency noise

Qian Li; Bo Yang; Xiaodong Wang; Lidong Liu; Chuanming Liu; Junbo Su

doi:10.1117/12.2505904

7 November 2018 Neural network non-uniformity correction for eliminating low frequency noise

Qian Li, Bo Yang, Xiaodong Wang, Lidong Liu, Chuanming Liu, Junbo Su

Proceedings Volume 10832, Fifth Conference on Frontiers in Optical Imaging Technology and Applications; 1083207 (2018) https://doi.org/10.1117/12.2505904
Event: Fifth Conference on Frontiers in Optical Imaging Technology and Applications, 2018, Changchun, China

Abstract

Classic neural network algorithm can not correct the low frequency noise,and it is easy to appearance ghost artifact phenomenon. In this paper,we present a new improved algorithm which can eliminate the low frequency noise and ghost artifact.We add a learning layer which can preprocess to the input layer.In this layer, use the Gaussian filter folding the down sampling image and resampling to the image so that the new input image have global correlation, it could assure the accuracy of the estimate real scene image,this can help us to eliminate the low frequency noise in the input image. Then we judged the frame motion by the edge detection, setting a threshold to compare with the MSE of the edge frame difference.Considering the result of the comparison,we use the difference step size to update the gain and offset parameters. Experimental results show that the new algorithm can improve image quality effectively

Citation Download Citation

Qian Li, Bo Yang, Xiaodong Wang, Lidong Liu, Chuanming Liu, and Junbo Su "Neural network non-uniformity correction for eliminating low frequency noise", Proc. SPIE 10832, Fifth Conference on Frontiers in Optical Imaging Technology and Applications, 1083207 (7 November 2018); https://doi.org/10.1117/12.2505904

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