Newsgroups: sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!news.sei.cmu.edu!cis.ohio-state.edu!math.ohio-state.edu!cs.utexas.edu!convex!seas.smu.edu!ksk
From: ksk@seas.smu.edu (Koduri S.)
Subject: Re: Image Registration Transformation Methods (File 1)
Message-ID: <1994Oct7.183953.10445@seas.smu.edu>
Sender: news@seas.smu.edu (USENET News System)
Nntp-Posting-Host: quick_f.seas.smu.edu
Organization: SMU - School of Engineering and Applied Science
References: <36seq0$suj@canopus.cc.umanitoba.ca> <1994Oct6.164346.21295@seas.smu.edu> <371rkf$78g@onramp.arc.nasa.gov>
Date: Fri, 7 Oct 1994 18:39:53 GMT
Lines: 19

In article <371rkf$78g@onramp.arc.nasa.gov>,
Kevin Montgomery <kevin@biocomp.arc.nasa.gov> wrote:
>
>FYI, I've been working on the next harder step, which is registration of 
>objects without any fiducial markers.  One context where this occurs is 
>in biomedical imaging- on a section of tissue, sometimes there are no 
>landmarks that you can 100% rely upon.  Then it becomes an error minimization
>problem and takes alot of crunching, but we're finding that it is completely 
>tractable nonetheless...
>						-Kevin
>


   KEVIN,

     THAT IS INTERESTING. DO YOU HAVE ANY PUBLICATIONS/REFERENCES?
     I WANT TO LOOK AT THE IMAGES WITHOUT ANY FIDUCIAL MARKS.

     -KODURI
