Title :
Hierarchical Bayesian image restoration from partially-known blurs
Author :
Mesarovic, Vladimir ; Galatsanos, Nikolas ; Molina, Rafael ; Katsaggelos, Aggelos
Author_Institution :
Crystal Semicond. Corp., Austin, TX, USA
Abstract :
A number of restoration filters have been proposed for the restoration problem from partially-known blurs. Previously we proposed the regularized constrained least-squares filter (RCTLS) and showed that it has a number of advantages over previous ones (Mesarovic et al. 1995). However, the problem of estimating the parameters that define the RCTLS filter has not yet been addressed. In this paper we propose a two-step algorithm based on the hierarchical Bayesian approach to simultaneously restore the image and estimate the parameters of the RCTLS restoration filter. The algorithm is derived in the DFT domain; thus, it is very efficient even for very large images
Keywords :
Bayes methods; digital filters; discrete Fourier transforms; image enhancement; image restoration; least squares approximations; parameter estimation; DFT; RCTLS filter; hierarchical Bayesian image restoration; partially-known blurs; regularized constrained least-squares filter; restoration filters; two-step algorithm; very large images; Algorithm design and analysis; Bayesian methods; Covariance matrix; Discrete Fourier transforms; Image restoration; Iterative algorithms; Mean square error methods; Nonlinear filters; Parameter estimation; Tomography;
Conference_Titel :
Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-4428-6
DOI :
10.1109/ICASSP.1998.678133
