Blind deconvolution of images using optimal sparse representations

Author

Bronstein, Michael M. ; Bronstein, Alexander M. ; Zibulevsky, Michael ; Zeevi, Yehoshua Y.

Author_Institution

Dept. of Comput. Sci., Technion - Israel Inst. of Technol., Haifa, Israel

Volume

14

Issue

6

fYear

2005

fDate

6/1/2005 12:00:00 AM

Firstpage

726

Lastpage

736

Abstract

The relative Newton algorithm, previously proposed for quasi-maximum likelihood blind source separation and blind deconvolution of one-dimensional signals is generalized for blind deconvolution of images. Smooth approximation of the absolute value is used as the nonlinear term for sparse sources. In addition, we propose a method of sparsification, which allows blind deconvolution of arbitrary sources, and show how to find optimal sparsifying transformations by supervised learning.

Keywords

Newton method; approximation theory; blind source separation; deconvolution; image representation; maximum likelihood estimation; optimisation; smoothing methods; Newton algorithm; blind deconvolution; image sparse representation; optimization; quasimaximum likelihood blind source separation; smooth approximation; Additive noise; Blind source separation; Deconvolution; Degradation; Image restoration; Image sensors; Kernel; Maximum likelihood estimation; Optical distortion; Sensor systems; Blind deconvolution; quasi-maximum likelihood; relative Newton optimization; sparse representations; Algorithms; Image Enhancement; Image Interpretation, Computer-Assisted; Information Storage and Retrieval; Pattern Recognition, Automated; Regression Analysis; Statistics as Topic;

fLanguage

English

Journal_Title

Image Processing, IEEE Transactions on

Publisher

ieee

ISSN

1057-7149

Type

jour

DOI

10.1109/TIP.2005.847322

Filename

1430762