Title :
On the use of discrete Laplacian operators in image restoration
Author :
Leung, C.M. ; Lu, W.-S.
Author_Institution :
Dept. of Electr. & Comput. Eng., Victoria Univ., BC, Canada
Abstract :
The role of the stabilizing functional in the Tikhonov (1964) regularization is as crucial as that of the regularization parameter. However, a 3×3 discrete differential operator is usually used to generate the stabilizing functional. It is demonstrated that the restoration quality is practically independent of the size of the generating stabilizing operator. As a result, the choice of small size of the generating stabilizing operator is justified. Moreover, this observation leads to the advantage that spatial-domain implemented filters can be designed within a reasonable small order
Keywords :
Laplace equations; differential equations; filtering theory; functional equations; image restoration; stability; discrete Laplacian operators; discrete differential operator; generating stabilizing operator; image restoration; regularization parameter; restoration quality; spatial-domain filters; stabilizing functional; stabilizing operator; Additive white noise; Degradation; Discrete Fourier transforms; Filters; Frequency estimation; Image restoration; Laplace equations; Layout; Signal restoration; Signal to noise ratio;
Conference_Titel :
Communications, Computers, and Signal Processing, 1995. Proceedings., IEEE Pacific Rim Conference on
Conference_Location :
Victoria, BC
Print_ISBN :
0-7803-2553-2
DOI :
10.1109/PACRIM.1995.519556
