A nonlinear total variation-based denoising method with two regularization parameters

Corina S Drapaca

doi:10.1109/TBME.2008.2011561

A nonlinear total variation-based denoising method with two regularization parameters

IEEE Trans Biomed Eng. 2009 Mar;56(3):582-6. doi: 10.1109/TBME.2008.2011561. Epub 2009 Jan 23.

Author

Corina S Drapaca¹

Affiliation

¹ Department of Engineering Science and Mechanics, Pennsylvania State University, University Park, PA 16802, USA. csd12@psu.edu

PMID: 19174344
DOI: 10.1109/TBME.2008.2011561

Abstract

The aim of the present paper is to study the effect of the regularization parameter used in the numerical implementation of the Rudin-Osher-Fatemi denoising model. By using two different regularization parameters in the numerical scheme of the Rudin-Osher-Fatemi model, we will show experimentally that when a particular relationship between the sizes of these parameters holds, the quality of the denoised image and the speed of convergence of the numerical scheme are both much improved in comparison with the classic numerical scheme of the Rudin-Osher-Fatemi model where only one regularization parameter is used.

Publication types

Comparative Study

MeSH terms

Algorithms*
Image Processing, Computer-Assisted / methods*
Models, Theoretical*
Phantoms, Imaging