H/sub /spl infin// deconvolution filter design and its application in image restoration

This paper addresses the design of an H/sub /spl infin// deconvolution filter and its application to image restoration. The proposed H/sub /spl infin// deconvolution filter has some advantages for image restoration in its capability of handling unknown boundary problems and spatially varying blurs. In this paper, the H/sub /spl infin// filter is compared with the inverse Wiener filter and a regularized restoration method. The experimental results show that the H/sub /spl infin// filter deals with the unknown boundary problem better than the Wiener filter. Compared with the regularization method, it gives a sharper restored image, especially, when the original image contains many details.

[1]  Uri Shaked,et al.  A frequency domain approach to the problems of H∞-minimum error state estimation and deconvolution , 1992, IEEE Trans. Signal Process..

[2]  Reginald L. Lagendijk,et al.  Regularized iterative image restoration in a weighted Hilbert space , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3]  Xianggang Yu,et al.  Reduced order H/sub /spl infin// filter design for discrete time-varying systems , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[4]  Masayuki Fujita,et al.  Finite horizon discrete-time H/sub /spl infin// filter with application to an active vision system , 1993, Proceedings of 32nd IEEE Conference on Decision and Control.

[5]  R. Mersereau,et al.  Optimal estimation of the regularization parameter and stabilizing functional for regularized image restoration , 1990 .

[6]  David J. N. Limebeer,et al.  Linear Robust Control , 1994 .

[7]  Jan Biemond,et al.  Boundary value problem in image restoration , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  Michael J. Grimble,et al.  H ∞ optimal multichannel linear deconvolution filters, predictors and smoothers , 1996 .