Plugin GIMP : Fourier Transform

I.   What it does

 It does a direct and reverse Fourier Transform.
 It allows you to work in the frequency domain.
 For instance, it can be used to remove moiré patterns from
images scanned from books. (See README.Moire)

II.  Use 

 It adds menu items :
  Filters/Generic/Foward FFT
  Filters/Generic/Inverse FFT

III. Installation

 Linux:
   make
   make install

   You will need the fftw3 package, and the development packages 
    of gimp, fftw3, and glib.

 Win32: 
 
  Binaries for win32 are provided (in a separate package), just copy the files
  in the plugins directory of :
   your personal gimp directory (ex: .gimp-2.2\plug-ins), 
   or in the global directory (C:\Program Files\GIMP-2.2\lib\gimp\2.0\plug-ins)
   
   If you don't have fftw3 installed on your system, you must download 
   the provided fftw3.dll, and copy it where you stored the plugin.

   You can compile it with the free environment DevCPP (mingw compiler). 
   You will need several DevPak : fftw3, GTK and gimp-dev (available on 
   devpaks.org). gimp-dev is designed to work with binaries version of 

History :
 v0.1.1 : First release of this plugin
 v0.1.2 : BugFixes by Mogens Kjaer <[email protected]>, May 5, 2002 
 v0.1.3 : Converted to Gimp 2.0 (dirty conversion)
 v0.2.0 : Many improvements from Mogens Kjaer <[email protected]>, Mar 16, 2005
              * Moved to gimp-2.2
              * Handles RGB and grayscale images
              * Scale factors stored as parasite information
              * Columns are swapped
 v0.3.0 : Great Improvement from Alex Fernández with dynamic boosting :
              * Dynamic boosted normalization : 
                    fft/inverse loss of quality is now un-noticeable 
              * Removed the need of parasite information
 v0.3.1 : Zero initialize padding (patch provided by Rene Rebe)
 v0.3.2 : GPL, Fixed Makefile with pkg-config
 v0.4.0 : Patch by Edgar Bonet :
             * Reordered the data in a more natural way
             * No Fourier coefficient is lost
 
Many thanks to Mogens Kjaer, Alex Fernández, Rene Rebe and Edgar Bonet 
for their contributions.

-- 
Rémi Peyronnet
[email protected]
http://www.via.ecp.fr/~remi