Abstract
The convergence properties of the iterative revision method for image restoration are examined when data are available only in restricted multiple regions. First, it is shown that this image restoration process is equivalent to solving a linear equation. Then a method of constructing an optimum observation system is shown that minimizes the mean-square error of the image obtained by stopping the iterative revisions at a finite number. Concrete optimum systems are derived for two cases, (i) when the sum of the widths of the restricted regions can be changed and (ii) when the sum of the widths of the restricted regions is fixed.
© 1982 Optical Society of America
Full Article | PDF ArticleMore Like This
K. Matsuoka, T. Shigematsu, Y. Ichioka, and T. Suzuki
Appl. Opt. 21(24) 4493-4499 (1982)
Junji Maeda and Kazumi Murata
Appl. Opt. 21(12) 2199-2204 (1982)
R. Mammone and G. Eichmann
Appl. Opt. 21(3) 496-501 (1982)