Multichannel image identification and restoration based on the EM algorithm and cross-validation

Aggelos K. Katsaggelos; Nikolas P. Galatsanos; Kuen-Tsair Lay; Wenwu Zhu

doi:10.1117/12.139009

29 December 1992 Multichannel image identification and restoration based on the EM algorithm and cross-validation

Aggelos K. Katsaggelos, Nikolas P. Galatsanos, Kuen-Tsair Lay, Wenwu Zhu

Proceedings Volume 1767, Inverse Problems in Scattering and Imaging; (1992) https://doi.org/10.1117/12.139009
Event: San Diego '92, 1992, San Diego, CA, United States

Abstract

In this paper we address some of the main shortcomings of multi-channel (MC) linear restoration filters. The problem of restoring a MC image and simultaneously estimating the MC power spectrum of the image and the noise, required by linear minimum mean squared error (LMMSE) filters is investigated, using the expectation-maximization (EM) algorithm. Second, the problem of estimating, the regularization parameters and operator, required by regularized least-squares (RLS) MC restoration filters is investigated using the cross-validation (CV) function. Furthermore, a novel representation of MC signal processing is introduced. This notation leads to a more natural extension of single-channel (SC) signal processing algorithms to the MC case and yields a new class of matrices which we call semi-block- circulant (SBC) matrices. The properties of these matrices are examined and a family of new efficient algorithms is developed for the computation of the MC EM and CV functions.

Citation Download Citation

Aggelos K. Katsaggelos, Nikolas P. Galatsanos, Kuen-Tsair Lay, and Wenwu Zhu "Multichannel image identification and restoration based on the EM algorithm and cross-validation", Proc. SPIE 1767, Inverse Problems in Scattering and Imaging, (29 December 1992); https://doi.org/10.1117/12.139009

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