Re: FFT on a stack ( automatization)

Posted by Michael Schmid on Mar 02, 2009; 9:06am
URL: http://imagej.273.s1.nabble.com/FFT-on-a-stack-automatization-tp3693511p3693512.html

Hi Guenther,

Process>FFT>Custom filter works on stacks.

Take our FFT, make a mask from it that is 255 where wou want to pass  
the frequencies and 0 where you want to block it (essentially the  
same as you would do for normal filtering, but you have to change the  
values both inside *and* outside of your roi). Then smooth it  
slightly (3 times 'smooth' or Gaussian Blur with a small radius to  
aviod ringing artifacts).

In contrast to normal 'Inverse FFT', for 'Custom Filter' you have to  
take care about creating a symmetric filter, it won't be made  
symmetric with respect to the origin.

Michael
________________________________________________________________

On 2 Mar 2009, at 07:45, Guenter Giese wrote:

> Dear list members,
>
> I have stacks of images with darkened regions (shadow-like stripes)  
> running
> through (two main directions with angles only some degrees to  
> orthogonal
> directions). On a single image, I run FFT, edit the power spectrum  
> (clearing
> of a polygonal ROI), and then do inverse FFT.
>
> Is it possible to do automated FFT/ROI clearing/inverse FFT on a  
> whole stack
> using a polygonal (predefined?) ROI for editing the power spectrum?  
> Perhaps
> as a macro?
>
> Best,
>
> Guenter
>
> ------------------------------------------
> Dr. Guenter Giese
> Light Microscopy Facility Manager
> Dept. of Biomedical Optics
> MPI fuer Medizinische Forschung Jahnstr. 29
> D-69120 Heidelberg, Germany
> Phone [phone] (Fax: -325)
> e-mail: [hidden email]