Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83 SMI; site ur-laser.uucp
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!genrad!panda!talcott!harvard!seismo!rochester!ur-laser!nitin
From: nitin@ur-laser.uucp (Nitin Sampat)
Newsgroups: net.graphics
Subject: FFT of image in sections..
Message-ID: <298@ur-laser.uucp>
Date: Mon, 19-Aug-85 11:18:51 EDT
Article-I.D.: ur-laser.298
Posted: Mon Aug 19 11:18:51 1985
Date-Received: Sat, 24-Aug-85 15:15:46 EDT
Organization: Lab for Laser Energetics, Univ. of Rochester
Lines: 27

From what I have reading regarding this article since I posted it, it
seems to me that I may not have communicated my question clearly enough.
Carl Lowenstein and Shep Seigel mention that taking an FFT does involve
doing it in "sections" or representing a two dimensional string of 
numbers as a one dimensional string etc..All that is fine but my question
did not refer to the FFT algorithm itself or how it manifests itself(although
that would be an interesting discussion in itself) in "sections". 

I am saying that once we have a certain FFT algorithm that we start using,
we may be limited by the core memory depending on the computer we are 
using. If we have a megabyte image and we try an FFT on it, most computers
will start paging, copying to and fro from disk since they can't fit the entire
image in core at one time.  However, they can process a smaller image almost
instantaneously(that is an image small enough to get complete processing in
core without any significant paging).  Once we determine what the physical limitof our computer is, we will know the smallest image we can process FAST.

Given this information, my question is, can we now process the megabyte image
in such sections and get any increase in speed. Also, does linear system
theory allow such a process.. because an FFT of a part is NOT the FFT of the
whole image. 

I dont think we can take a 64 X 64 image ( say ) , break it up
into 16 X 16 sub-images, FFT them and put these FFT'd sub-images together
to say that this is the FFT of the original 64 X 64 image, or can we ?

				nitin
				{allegra,seismo}!rochester!ur-laser.uucp