We propose an information-theoretic variational filter for image denoising. It is a result of minimizing a functional subject to some noise constraints, and takes a hybrid form of a negentropy variational integral for small gradient magnitudes and a total variational integral for large gradient magnitudes. The core idea behind this approach is to use geometric insight in helping to construct regularizing functionals and avoiding a subjective choice of a prior in maximum a posteriori estimation. Illustrative experimental results demonstrate a much improved performance of the approach in the presence of Gaussian and heavy-tailed noise.
This article is part of the series Nonlinear Signal and Image Processing - Part II.
A Nonlinear Entropic Variational Model for Image Filtering
1 Concordia Institute for Information Systems Engineering, Concordia University, Montréal, Quebec H3G 1T7, Canada
2 Department of Electrical and Computer Engineering, North Carolina State University, Raleigh, NC 27695-7911, USA
3 Ariana Research Group, INRIA/I3S, BP 93, Sophia Antipolis Cedex, 06902, France
EURASIP Journal on Advances in Signal Processing 2004, 2004:540425 doi:10.1155/S1110865704407197
The electronic version of this article is the complete one and can be found online at: http://asp.eurasipjournals.com/content/2004/16/540425
|Received:||12 August 2003|
|Revisions received:||8 June 2004|
|Published:||2 December 2004|
© 2004 Ben Hamza et al.