Abstract

In this paper we report some experimental results of a direct algorithm for the solution of digital image restoration problems. The solution is obtained directly from the system of linear equations that result from the discretization of the Fredholm integral equation of the first kind. This algorithm uses a simple regularized least-squares technique. Also, the regularization parameter for the optimum solution is calculated by a direct (noniterative) method. A computer simulated example using both space invariant and space variant spatially separable point spread functions is presented. We show that this method compares favorably with other known direct methods.

© 1981 Optical Society of America

Image restoration for space invariant pointspread functions

N. N. Abdelmalek, T. Kasvand, and J. P. Croteau
Appl. Opt. 19(7) 1184-1189 (1980)

Digital image restoration using quadratic programming

N. N. Abdelmalek and T. Kasvand
Appl. Opt. 19(19) 3407-3415 (1980)

Image restoration by an iterative regularized pseudoinverse method

Junji Maeda and Kazumi Murata
Appl. Opt. 23(6) 857-861 (1984)

Constrained least-squares image restoration: an improved computational scheme

S. S. Reddi
Appl. Opt. 17(15) 2340-2341 (1978)

Iterative image restoration considering visibility

Yoshiki Ichioka and Nobuyoshi Nakajima
J. Opt. Soc. Am. 71(8) 983-988 (1981)