An improved neural network nonuniformity correction for IRFPA

Zhenguo Liu; Xiaomei Hu; Jin Lu

doi:10.1117/12.836535

4 August 2009 An improved neural network nonuniformity correction for IRFPA

Zhenguo Liu, Xiaomei Hu, Jin Lu

Proceedings Volume 7383, International Symposium on Photoelectronic Detection and Imaging 2009: Advances in Infrared Imaging and Applications; 738330 (2009) https://doi.org/10.1117/12.836535
Event: International Symposium on Photoelectronic Detection and Imaging 2009, 2009, Beijing, China

Abstract

The non-uniformity correction (NUC) is a key part that affects the image quality of IR imaging systems. In this paper, the NUC technique for staring IR imaging system is studied based on the research of the response characteristic and the noise components of IRFPA. Especially, the improvement of the traditional neural network correction method is also studied. According to the theory of neural network, we analyzed the reasons that cause the defects of traditional neural network correction, which are the difficulty for choosing the length of study-step, poor performance of the offset nonuniformity correction and the side-effect of "ghosting", then found methods to improve them. So the correction ability and the applicability of the improved NUC method are enhanced. To test the effect of the new NUC method, we implemented the arithmetic on a hardware platform using the Digital Signal Processor TS201, and made experiments in real terrestrial scene and offshore scene. The results prove that the improved neural network correction arithmetic is effective.

Citation Download Citation

Zhenguo Liu, Xiaomei Hu, and Jin Lu "An improved neural network nonuniformity correction for IRFPA", Proc. SPIE 7383, International Symposium on Photoelectronic Detection and Imaging 2009: Advances in Infrared Imaging and Applications, 738330 (4 August 2009); https://doi.org/10.1117/12.836535

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