MathGroup Archive 2006

[Date Index] [Thread Index] [Author Index]

Search the Archive

Gabor wavelet decomposition

  • To: mathgroup at smc.vnet.net
  • Subject: [mg64705] Gabor wavelet decomposition
  • From: Gareth Russell <russell at njit.edu>
  • Date: Tue, 28 Feb 2006 01:49:37 -0500 (EST)
  • Sender: owner-wri-mathgroup at wolfram.com

Hi Group,

I am going some image processing using the Wavelet Explorer toolbox. 
The toolbox uses the Least Asymmetric version of Daubechies as its 
example for 2D decomposition, but I would like to compare the results 
to the commonly-used Gabor filter, which is NOT part of the Toolbox.

I know what the Gabor function is, but am not clear whether I can 
implement this in such a way that I can use the other functions of the 
Toolbox. The Gabor function is explicitly a 2D function (rather than 
the product of a 1D function applied to x and y coordinates), so the 
Toolbox input format, in which the chosen filter is provided as a 1D 
list of coefficients, would seem not to work.

Please correct me if I'm wrong. If I'm right, does anyone know of any 
existing Mathematica code that does decomposition using Gabor wavelets? 
People have written code for a variety of other platforms, so before I 
try to adapt these on my own, I thought I would see what's available 
for Mathematica. I'm sure someone must have done it, but a search of 
the usual sources (MathGroup archives, etc.) comes up empty.

Thanks,

Gareth Russell
Rutgers University/NJIT


  • Prev by Date: code efficiency
  • Next by Date: Re: Limit
  • Previous by thread: code efficiency
  • Next by thread: Strange problem with InverseSeries