Blind Bayesian restoration of adaptive optics telescope images using generalized Gaussian Markov random field models

Brian D. Jeffs; Julian C. Christou

doi:10.1117/12.321721

11 September 1998 Blind Bayesian restoration of adaptive optics telescope images using generalized Gaussian Markov random field models

Brian D. Jeffs, Julian C. Christou

Author Affiliations +

Proceedings Volume 3353, Adaptive Optical System Technologies; (1998) https://doi.org/10.1117/12.321721
Event: Astronomical Telescopes and Instrumentation, 1998, Kona, HI, United States

Abstract

This paper addresses post processing for resolution enhancement of sequences of short exposure adaptive optics (AO) images of space objects. The unknown residual blur is removed using Bayesian maximum a posteriori blind image restoration techniques. In the problem formulation, both the true image and the unknown blur psf's are represented by the flexible generalized Gaussian Markov random field (GGMRF) model. The GGMRF probability density function provides a natural mechanism for expressing available prior information about the image and blur. Incorporating such prior knowledge in the deconvolution optimization is crucial for the success of blind restoration algorithms. For example, space objects often contain sharp edge boundaries and geometric structures, while the residual blur psf in the corresponding partially corrected AO image is spectrally band limited, and exhibits while the residual blur psf in the corresponding partially corrected AO image is spectrally band limited, and exhibits smoothed, random , texture-like features on a peaked central core. By properly choosing parameters, GGMRF models can accurately represent both the blur psf and the object, and serve to regularize the deconvolution problem. These two GGMRF models also serve as discriminator functions to separate blur and object in the solution. Algorithm performance is demonstrated with examples from synthetic AO images. Results indicate significant resolution enhancement when applied to partially corrected AO images. An efficient computational algorithm is described.

Citation Download Citation

Brian D. Jeffs and Julian C. Christou "Blind Bayesian restoration of adaptive optics telescope images using generalized Gaussian Markov random field models", Proc. SPIE 3353, Adaptive Optical System Technologies, (11 September 1998); https://doi.org/10.1117/12.321721

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