[Insight-users] about VersorRigid3D

[Li, George (NIH/NCI)]

Yixun:
 
You have to show what the result looks like and what your expectation is
before your question can be addressed. "the result is not good" is vague.
 
George
 
 
-----Original Message-----
From: Yixun Liu [mailto:yxliu at fudan.edu.cn] 
Sent: Wednesday, January 12, 2005 1:29 AM
To: insight-users at itk.org
Subject: [Insight-users] about VersorRigid3D



Hi,
 
I want to register two 3D CT datasets, which came from one patient and were
scanned at different time.
 
the size of the two datasets both are: 512*512*32,  spacing: 1.36*1.36*5.0
 
The metric I use is: MattesMutualInformationImageToImageMetric,
The Transform is: VersorRigid3D
The Optimizor: VersorRigid3DTransformOptimizer
 
the paramaters are:
METRIC:
  metric->SetNumberOfHistogramBins( 20 );
  metric->SetNumberOfSpatialSamples( 512*512*32 / 100);
 
 
OPTIMIZER
  optimizerScales[0] = 1.0;
  optimizerScales[1] = 1.0;
  optimizerScales[2] = 1.0;
  optimizerScales[3] = 1/100000;
  optimizerScales[4] = 1/100000;
  optimizerScales[5] = 1/100000;
  optimizer->SetScales( optimizerScales );
 
  optimizer->SetMaximumStepLength( 0.2000  ); 
  optimizer->SetMinimumStepLength( 0.0001 );
  optimizer->SetNumberOfIterations( 200 );

However, the result is not good. I do not intend to use the
MeanSquaresImageToImageMetric due to its slowness. The reason may be the
incorrect paramters. Could anybody tell me how to set the  correct
parameters?
 
Best regards
 




 
 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20050112/b90ba9bd/attachment.htm