Image Denoising Using Wavelets: Application in Medical Imaging - Advances in Heuristic Signal Processing and Applications - page 281

Chapter 13
Image Denoising Using Wavelets:
Application in Medical Imaging

Abdeldjalil Ouahabi

Abstract Medical images obtained from MRI are the most common tool for di-
agnosis in Medicine. These images are often affected by random noise arising in
the image acquisition process. Hence, noise removal is essential in medical imaging
applications in order to enhance and recover fine details that may be hidden in the
data.
A common approach for image denoising is to convert a noisy image into a trans-
form domain such as the wavelet and contourlet domain, and then compare the trans-
form coefficients with a fixed or adapted threshold. The underlying idea is that the
useful signal can be described by a small number of coefficients of high-amplitude
wavelets, and that the noise is spread across all coefficients. In fact, the wavelet rep-
resentation naturally compresses the essential information in a signal into relatively
few, large coefficients, which represent image details at different resolution scales.
In this chapter, we review recent wavelet denoising techniques for medical ultra-
sound and for magnetic resonance images and discuss their performances in terms
of SNR (or PSNR) and visual aspects of image quality. However, image denoising
using wavelet-based multiresolution analysis requires a delicate compromise be-
tween noise reduction and preserving significant image details. Hence, in practical
applications, we will often simplify the theory using heuristics, when this leads to
algorithms with lower complexity or higher flexibility.

13.1 Introduction to Multiresolution Analysis

This introductory section recalls the theory of wavelets and multiresolution analysis
based on the discrete wavelet transform. The Mallat algorithm for 1D and 2D signals
as a tool for concrete implementation of these concepts is shown at the end of this
section.

B

A. Ouahabi ( )

Polytech Tours, Tours University, 7 Avenue Marcel Dassault, 37200 Tours, France
e-mail: [e-mail]