Newsgroups: sci.image.processing,sci.math.num-analysis,comp.dsp,alt.image.medical
Path: cantaloupe.srv.cs.cmu.edu!rochester!rutgers!news.iag.net!usenet.eel.ufl.edu!hookup!uwm.edu!msunews!harbinger.cc.monash.edu.au!news.cs.su.oz.au!metro!cassius!adrianw
From: adrianw@cassius (Adrian Whichello)
Subject: Re: FFT of a mask?
Message-ID: <D9FFJt.26y@ucc.su.OZ.AU>
Sender: news@ucc.su.OZ.AU
Nntp-Posting-Host: cassius.ee.su.oz.au
Organization: Electrical Engineering, University of Sydney
References: <3psn4i$gi@lyra.csx.cam.ac.uk> <3pv0tj$1ll@ucthpx.uct.ac.za>
Distribution: inet
Date: Wed, 31 May 1995 05:07:04 GMT
Lines: 49
Xref: glinda.oz.cs.cmu.edu sci.image.processing:14941 sci.math.num-analysis:21122 comp.dsp:19878

In article <3pv0tj$1ll@ucthpx.uct.ac.za> greg@dip1.ee.uct.ac.za writes:
>In article <3psn4i$gi@lyra.csx.cam.ac.uk>, tph1001@cus.cam.ac.uk (T.P Harte) writes:
>>
>>I want to convolve a 256 x 256 image with a 3 x 3 mask: this is a convolution
>>in the FFT sense, i.e., I want to FFT the image and FFT the mask and then
>>multiply them together.
>>
>>How do I get the 3 x 3 mask into a suitable format to FFT it? I presume that I
>>will have to zero-pad it and *somehow* make it into a 256 x 256 image prior to
>>FFTing and multiply; it will have to be a power of 2 anyhow.
>>
>>I'm just not so sure how *exactly* you go about doing this. Any suggestions?
>
>
>Thomas,
>
>Have a look at:
>
>@ARTICLE{Anuta70,
>        AUTHOR             = {Paul E. Anuta},
>        JOURNAL            = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
>        MONTH              = oct,
>        NUMBER             = {4},
>        PAGES              = {353-68},
>        TITLE              = {Spatial Registration of Multispectral and Multitemporal Digital Imagery Using Fast Fourier Transform Techniques},
>        VOLUME             = {GE-8},
>        YEAR               = {1970}
>}
>
>
>-- 
>Greg Cox                      greg@dip1.ee.uct.ac.za
>Image Processing Labs
>Electrical Engineering 
>University of Cape Town, South Africa 
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>


NB that should be
        JOURNAL            = {IEEE Transactions on Geoscience Electronics}, 

Pattern Analysis and Machine Intelligence only started in 1979.

Adrian Whichello                           Phone: +61 2 351 4824
Imaging Science and Engineering Laboratory  Home: +61 47 588 319 
Sydney University Electrical Engineering     Fax: +61 2 351 3847
Australia                                  Email: adrianw@ee.su.oz.au
