Premium
Qualified Thresholds forWavelet Shrinkage
Author(s) -
Boese F. G.
Publication year - 2005
Publication title -
pamm
Language(s) - English
Resource type - Journals
ISSN - 1617-7061
DOI - 10.1002/pamm.200510334
Subject(s) - shrinkage , wavelet , noise (video) , statistics , noise reduction , mathematics , reduction (mathematics) , pattern recognition (psychology) , shrinkage estimator , artificial intelligence , computer science , geometry , mean squared error , image (mathematics) , minimum variance unbiased estimator , bias of an estimator
One of many method for noise reduction in measured data is the wavelet‐based Wavelet Shrinkage propagated by D. L. Donoho, I. M. Johnstone et al. since about 1990. A crucial step within this approach is the setting of the treshold which decides whether a wavelet coefficient is considered as noise or as data. We propose an one‐paramerter family of thresholds. The parameter is the probability of making a false decision. (© 2005 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)