Hi Luis,<br><br>As you know, sometimes the orientation information in the header file is not correct.<br>For a clean registration, I do this:<br>1) I set the origin of both images to (0,0,0)<br>2) by using a translation transform, I search the space to align the images.<br>
3) I set the center of rotation to the center of the fixed image.<br>4) using the some rotations, i calculate a similarity measure for choosing the best rotation.<br>5) Initialize the transform and then .....<br><br><br>Is it effective? or is there other ways to do that?<br>
<br>Best,<br>MJM<br><br><div class="gmail_quote">On Mon, Jan 4, 2010 at 12:40 AM, Luis Ibanez <span dir="ltr"><<a href="mailto:luis.ibanez@kitware.com">luis.ibanez@kitware.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi MJM,<br>
<br>
The optimization process is very unlikely to correct<br>
for rotations larger than 20 degrees.<br>
<br>
If you know that your images are misaligned by<br>
more than 20 degrees, then your best option is to<br>
provide a good initialization for the transform.<br>
<br>
For example:<br>
via the use of the LandmarkTransformInitializer class.<br>
<br>
<br>
<br>
Beware, however, that it is common for medical images<br>
to be acquired at different orientations and for those<br>
orientations to be annotated as part of the image header,<br>
particularly if you are using DICOM. Therefore, before<br>
you conclude that your images are off by as much as 180<br>
degrees, you should review if there is any orientation<br>
information in the header of those images.<br>
<br>
One good test is to open both images into Slicer.<br>
Slicer is one of the few applications that will display<br>
correctly the relative positions of medical images,<br>
including both origin and orientation.<br>
<br>
You can get Slicer from:<br>
<a href="http://www.slicer.org/" target="_blank">http://www.slicer.org/</a><br>
<br>
<br>
Regards,<br>
<br>
<br>
Luis<br>
<br>
<br>
<br>
-----------------------------------------------------------------<br>
<div><div></div><div class="h5">On Fri, Dec 25, 2009 at 8:30 AM, Mostafa Jabarouti Moghaddam<br>
<<a href="mailto:zavata.afnan@gmail.com">zavata.afnan@gmail.com</a>> wrote:<br>
> Hello all,<br>
><br>
> If the orientations of the fixed image and moving image are not same, How<br>
> can we expand the parameter space? for example, we could change the rotation<br>
> parameter range between [-180,180] in each direction!<br>
> I searched most of the e-mails in ITK, but i just found out that the best<br>
> way is choosing the proper initial transform. Unfortunately, choosing the<br>
> proper initial transform needs an extensive computation!<br>
><br>
><br>
> Best,<br>
> MJM<br>
><br>
</div></div>> _____________________________________<br>
> Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
><br>
> Visit other Kitware open-source projects at<br>
> <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
><br>
> Kitware offers ITK Training Courses, for more information visit:<br>
> <a href="http://www.kitware.com/products/protraining.html" target="_blank">http://www.kitware.com/products/protraining.html</a><br>
><br>
> Please keep messages on-topic and check the ITK FAQ at:<br>
> <a href="http://www.itk.org/Wiki/ITK_FAQ" target="_blank">http://www.itk.org/Wiki/ITK_FAQ</a><br>
><br>
> Follow this link to subscribe/unsubscribe:<br>
> <a href="http://www.itk.org/mailman/listinfo/insight-users" target="_blank">http://www.itk.org/mailman/listinfo/insight-users</a><br>
><br>
><br>
</blockquote></div><br>