stack alignment / registration

[Michael Calhoun]

Hi.  Seems like this problem has been around for awhile, but I couldn't
find a solution in the software/plug-ins.  We would like to register 3D
imagery taken over multiple days, with the first stack as a reference.  
We have readily identifiable fiduciary points, and only need a linear
transform (but it needs to be able to translate/rotate in full 3D, not
slice by slice as the data also 'tilts').  Any free (!) suggestions?  
Thanks & cheers,

-Michael

_______________________________________________________
Michael E. Calhoun, Ph.D.
Department of Cellular Neurology
Hertie-Institute for Clinical Brain Research
University of Tübingen
Otfried-Müller-Str. 27
D-72076 Tübingen
Germany

Phone  xx49-7071-29-87607
Fax  xx49-7071-29-4521
http://www.hih-tuebingen.de/imaging.html

[Abramoff-Michael]

Yes, MI 3D registration has been around for awhile but to my knowledge it has not been implemented as a plugin (lack of people needing it). The BIJ library contains everything you might need including 3D joint histograms, minimization algorithms and 3D linear transformation. But you are on your own in putting it together and testing.

Sorry, Michael Abramoff

Michael D. Abràmoff, MD, PhD
Assistant Professor of Clinical Ophthalmology / Assistant Professor of Electrical and Computer Engineering
Department of Ophthalmology and Visual Sciences
University of Iowa Hospital and Clinics, PFP 11290C
200 Hawkins Drive Iowa City, IA 52242
USA

Tel: [phone]. Secretary Val Bell: [phone].
Skype michael-abramoff [hidden email]

 

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Michael Calhoun
Sent: Wednesday, July 27, 2005 3:11 PM
To: [hidden email]
Subject: stack alignment / registration

Hi.  Seems like this problem has been around for awhile, but I couldn't find a solution in the software/plug-ins.  We would like to register 3D imagery taken over multiple days, with the first stack as a reference.  
We have readily identifiable fiduciary points, and only need a linear transform (but it needs to be able to translate/rotate in full 3D, not slice by slice as the data also 'tilts').  Any free (!) suggestions?  
Thanks & cheers,

-Michael

_______________________________________________________
Michael E. Calhoun, Ph.D.
Department of Cellular Neurology
Hertie-Institute for Clinical Brain Research University of Tübingen Otfried-Müller-Str. 27
D-72076 Tübingen
Germany

Phone  xx49-7071-29-87607
Fax  xx49-7071-29-4521
http://www.hih-tuebingen.de/imaging.html

[J. Anthony Parker]

Michael,

Align3_TP, http://www.med.harvard.edu/JPNM/ij/plugins/Align3TP.html,  
allows the user to do 3D alignment as a sequence of 2D operations.  
The alignment is an affine transform.  It is primarily meant to be  
used for manual registration, but I started to play a bit with  
automatic registration including 3D.  Unfortunately, I got distracted  
by other tasks, and the automatic registration is not very user  
friendly.  The manual registrations is slightly more user friendly.

Tony
J. Anthony Parker, MD PhD
[hidden email]
tel [phone] fax -2185 comp -5275
page 617 632-7243 #31978
330 Brookline Ave, Boston, MA, 02215-5400


On Jul 27, 2005, at 4:11 PM, Michael Calhoun wrote:

> Hi.  Seems like this problem has been around for awhile, but I  
> couldn't find a solution in the software/plug-ins.  We would like  
> to register 3D imagery taken over multiple days, with the first  
> stack as a reference.  We have readily identifiable fiduciary  
> points, and only need a linear transform (but it needs to be able  
> to translate/rotate in full 3D, not slice by slice as the data also  
> 'tilts').  Any free (!) suggestions?  Thanks & cheers,
>
> -Michael
>
> _______________________________________________________
> Michael E. Calhoun, Ph.D.
> Department of Cellular Neurology
> Hertie-Institute for Clinical Brain Research
> University of Tübingen
> Otfried-Müller-Str. 27
> D-72076 Tübingen
> Germany
>
> Phone  xx49-7071-29-87607
> Fax  xx49-7071-29-4521
> http://www.hih-tuebingen.de/imaging.html
>