Wavelet Denoising Using a Conjointly Space and 2D Frequency Localized Filterbank

摘要

This paper proposes the use of a conjointly space-frequency well lo-calized separable 2D quadrature mirror filterbank to the application of wavelet image shrinkage. The novel space-frequency localization measure to determine optimality is the product of a filter's 2D variance in space and 2D frequency variance. The particle swarm optimization method is applied to determine perfect or near perfect reconstruction optimal 2D filterbanks. The experimentation and results shows that the proposed 2D optimized QMF is robust at reducing additive white Gaussian noise for a wide range of threshold parameters.