DocumentCode
329520
Title
Complexity-regularized image restoration
Author
Liu, Juan ; Moulin, Pierre
Author_Institution
Beckman Inst. for Adv. Sci. & Technol., Illinois Univ., Urbana, IL, USA
Volume
1
fYear
1998
fDate
4-7 Oct 1998
Firstpage
555
Abstract
We propose the use of complexity regularization in image restoration. This is a flexible estimation method which borrows from previous developments in nonparametric estimation theory. The regularized estimation problem is formulated in the wavelet domain and solved using a computationally efficient multiscale relaxation algorithm
Keywords
computational complexity; image restoration; maximum likelihood estimation; optimisation; smoothing methods; wavelet transforms; complexity-regularized image restoration; computationally efficient multiscale relaxation algorithm; maximum likelihood estimation; nonparametric estimation theory; nonquadratic smoothness penalities; optimization; regularized estimation problem; wavelet domain; AWGN; Additive white noise; Cost function; Estimation theory; Gaussian noise; Image restoration; Maximum likelihood estimation; Wavelet coefficients; Wavelet domain; Wavelet transforms;
fLanguage
English
Publisher
ieee
Conference_Titel
Image Processing, 1998. ICIP 98. Proceedings. 1998 International Conference on
Conference_Location
Chicago, IL
Print_ISBN
0-8186-8821-1
Type
conf
DOI
10.1109/ICIP.1998.723563
Filename
723563
Link To Document