Noise reduction by wavelet thresholding (Q5934759): Difference between revisions

Wavelet analysis is a modern and growing tool in today's mathematics. It has already found many applications. Among the most important ones are applications to image analysis and signal processing. One such application is the topic of this book. It deals with mathematical methods of noise reduction. The main idea is to use the sparsity of wavelet representations of signals and design the most effective threshold for its coefficients. The book starts with a short overview of Fourier and wavelet analysis, discusses the issues concerned with the minimum mean squared error thresholding and the universal threshold. It proposes the use of the generalized cross validation (GCV) function for fast threshold estimation. Next, the author exploits the structure of multiresolution analysis to design a level dependent threshold estimation with GCV. For the purpose of refining the denoised signals, a Bayesian analysis with multiscale Markov random field models is introduced. Finally, problems involving nonequispaced data are discussed. Since the book provides only a limited discussion of wavelet analysis, we would like to refer the reader interested in a more ``in-depth'' approach to e.g., \textit{E. Hernández} and \textit{G. Weiss} [A first course on wavelets, Studies in Advanced Mathematics (CRC Press, Boca Raton) (1996; Zbl 0885.42018)] and \textit{M. V. Wickerhauser} [Adapted wavelet analysis from theory to software (Wellesley) (1994; Zbl 0818.42011)]. Issues related to those discussed in the book may be found also in e.g., \textit{R. Coifman} and \textit{M. V. Wickerhauser} [Experiments with adapted wavelet de-noising for medical signals and images, in: Akay, Metin (ed.), Time-frequency and wavelets in biomedical signal processing (IEEE, New York) 323-346 (1998; Zbl 0917.92020)] and \textit{M. H. Benteftifa} and \textit{L. Kurz} [Analysis of variance in statistical image processing (Cambridge University Press, Cambridge) (1997)].