[Insight-users] Misregistration output difference between ITK 4.2 and 4.4

[Bradley Lowekamp]

Hello,

What is the y-axis?

Brad

On Nov 12, 2013, at 8:45 AM, Jagjeet Nain <jagjeet.nain at gmail.com> wrote:

> 
> 
> 
> Hi,
> I was trying to run an Image Registration program which uses ITK's Mattes Mutual Information method to compute misregistration between an image pair and is based on ../Examples/Registration/ImageRegistration4.cxx program.
> Then I linked this program with two different ITKs (3.2 & 4.4) and tried running two executable on ~200 image pairs to find misregistration between each pair.
> 
> I was expecting same output from two ITKs(running system is same, Linux 32 in both case) but I am bit puzzled to see that there are misregistration differences between two ITK versions, Also these differences are not negligible they are of the order of 0.1m in a pixel of size 3m. Attached are misregistration results from two ITKs.
> 
> I guess somebody might have done similar comparison between different ITK versions, what is their opinion? Are these differences expected?
> If differences between two ITKs are expected, then which one is better 3.2 or 4.4?
> Thanks.
> Any help or pointer in this respect is well appreciated.
> BR
> Jagjeet
> 
> <MisregistrationDifference3.2Vs4.4.jpg>_____________________________________
> Powered by www.kitware.com
> 
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
> 
> Kitware offers ITK Training Courses, for more information visit:
> http://www.kitware.com/products/protraining.php
> 
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
> 
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20131112/d5e12e5e/attachment.htm>