Blind deconvolution of images using optimal sparse representations

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

doi:10.1109/tip.2005.847322

Blind deconvolution of images using optimal sparse representations

IEEE Trans Image Process. 2005 Jun;14(6):726-36. doi: 10.1109/tip.2005.847322.

Authors

Michael M Bronstein¹, Alexander M Bronstein, Michael Zibulevsky, Yehoshua Y Zeevi

Affiliation

¹ Department of Computer Science, Technion-Israel Institute of Technology, Haifa 32000, Israel. bronstein@ieee.org

PMID: 15971772
DOI: 10.1109/tip.2005.847322

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.

Publication types

Evaluation Study
Research Support, Non-U.S. Gov't

MeSH terms

Algorithms*
Image Enhancement / methods*
Image Interpretation, Computer-Assisted / methods*
Information Storage and Retrieval / methods*
Pattern Recognition, Automated / methods
Regression Analysis
Statistics as Topic