<div><br>Hi Luis, </div>
<div> </div>
<div><font style="BACKGROUND-COLOR: #ffffff">thanks for the hints,</font></div>
<div><font style="BACKGROUND-COLOR: #ffffff"></font> </div>
<div><font style="BACKGROUND-COLOR: #ffffff" color="#000000">I am using Windows XP and Visual Studio 8. I used the release mode, and to initialize the transform that is perform with </font><font color="#800000" size="2">
<font style="BACKGROUND-COLOR: #ffffff" color="#000000">itkVersorRigid3DTransform and the initializer is </font><font size="2"><font style="BACKGROUND-COLOR: #ffffff" color="#000000">itkCenteredTransformInitializer.</font>
</font></font></div>
<div><font color="#800000" size="2"></font> </div>
<div><font size="2"><font color="#000000">For the optimizer<em> </em>I used <font size="2">itkVersorRigid3DTransformOptimizer</font></font></font></div>
<div><font size="2">
<p><font color="#000000">optimizer->SetMaximumStepLength( 4.00 ); </font></p>
<p><font color="#000000">optimizer->SetMinimumStepLength( 0.01 );</font></p></font></div>
<div><font color="#000000">I tried to fix these parameters, but maybe in the metrics I am using <font size="2">itkMeanSquaresImageToImageMetric I have to initialize some parameters to fast it up.</font></font></div>
<div><font size="2"></font> </div>
<div><font size="2">Laia</font></div>
<p><font color="#800000" size="2"></font> </p>
<div><br> </div>
<div><span class="gmail_quote">On 9/21/07, <b class="gmail_sendername">Luis Ibanez</b> <<a href="mailto:luis.ibanez@kitware.com">luis.ibanez@kitware.com</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid"><br>Hi Laia,<br><br><br> Welcome to ITK !<br><br><br>It should be possible to register, in about 5 minutes, images
<br>of the size that you describe, using modern desktop computers.<br><br><br>You may want to give it a try to fine tune a normal single-resolution<br>registration before you move into the multi-registration framework.<br>
<br><br><br>A couple of questions:<br><br><br> 1) What is your platform ? (compiler, operating system)<br><br><br> 2) Did you make sure to build ITK in Release mode ?<br> (a Debug build will be a lot slower).<br><br>
<br> 3) How are you initializing the Transform ?<br> (starting with a poorly initialized transform will<br> make very hard for the optimizer to find a solution<br> in a short time).<br><br><br> 4) Did you produced a log of the transform parameters
<br> at every iteration of the optimizer ?<br> This is very useful for fine tunning the parameters<br> of the registration.<br><br><br><br> Please let us know,<br><br><br> Thanks<br><br><br> Luis<br>
<br><br>-------------------<br>laia sales wrote:<br>> Hi All,<br>><br>> I am quite new in ITK and I am trying to use the Multi-Resolution<br>> registration from ITK to register my images.<br>><br>> I am currently using CT images from femur bones of aproximately
<br>> 250x250x500 with different resolutions of around 0.6x0.6x0.1 mm.<br>><br>> The Optimizer, i am using is the itkVersorRigid3DTransformOptimizer and<br>> the metric itkMeanSquaresImageToImageMetric.<br>>
<br>> I was firstly implementing the rigid registration with the code provided<br>> with the 3D registration, and it resulted pretty slow. Therefore, I<br>> wanted to try with the pyramidal registration to speed up the
<br>> registration, but it is not giving a faster process, but slowing down. I<br>> don't get well the parameters SetMaximumStepLength and<br>> SetMinimumStepLength. How these parameters can reduce the number of
<br>> iterations for each level and achieve a faster and optimal registration.<br>><br>> Does anybody use that kind of registration before and can give me some<br>> hints abut it?<br>><br>> Regards,<br>>
<br>><br></blockquote></div><br>