FWTN 1.0

Hannes Nickisch — Thu, 25 Mar 2010 16:58:24 -0000

The package includes a simple and standalone implementation of orthonormal wavelet transformations for D dimensional tensors in L levels. It is generic in the quadrature mirror filter, so any filter (Haar, Daubechies etc.) can be used. Runtime is O(n) with n being the number of elements of the tensor. The code is written in plain C (with inspirations from Wavelab); a Matlab/Octave mex wrapper as well as a demo example is provided.

In Matlab, you simply type the following to perform the transformation.

qmf = [1,1]/sqrt(2); % Haar Wavelet

L = 3; % # Levels in the pyramid

W = fwtn(X,L,qmf); % apply FWT, inverse: X = fwtn(W,L,qmf,1);

mloss.org FWTN

FWTN 1.0