Title :
Selection of natural scale in discrete wavelet domain using eigenvalues
Author :
Quddus, Azhar ; Gabbouj, Moncef
Author_Institution :
Tampere University of Technology (TUT), Signal Processing Laboratory, P.O. Box553, FIN-33101, Tampere, Finland
Abstract :
In this paper we present a novel technique for the selection of global natural scale from discrete wavelet transform. Here we define natural scale as the level associated with most prominent (dominant) eigenvalue. This technique is iterative and does not require full decomposition before finding the optimal wavelet level.
Keywords :
Discrete wavelet transforms; Eigenvalues and eigenfunctions; Filtering; Indexes; Matrix decomposition; Wavelet domain;
Conference_Titel :
Signal Processing Conference, 2000 10th European
Conference_Location :
Tampere, Finland
Print_ISBN :
978-952-1504-43-3
