[Insight-users] itkSymmetricEigenAnalysisImageFilter : see : "Vessel Enhancing Diffusion Filter" paper in the Insight Journal.

[Luis Ibanez]

Hi Ruben, Jon,

If you are planning to compute pixel-wise the Hessian of the image,
you may find useful to look at the code of the recent submission
to the Insight Journal by Andinet Enquobahrie:

"Vessel Enhancing Diffusion Filter"
http://www.insight-journal.org/InsightJournalManager/view_reviews.php?back=index.php&pubid=163

The diffusion filter presented in this paper compute the Hessian and
its Eigen Analysis on a pixel by pixel basis, without requiring the
large memory of the HessianRecursiveGaussian filter.  Note however,
that there is no smoothing implied here, so , it is up to you to prepare
the input image to make sure that it doesn't have high spatial frequency
noise.


    Regards,


       Luis


----------------------
Ruben Schilling wrote:
> Hi Jone,
> 
> If you are not going to implement it voxel-wise, but you would rather 
> use the built-in image to image filter, then this might be of help to you:
> 
> typedef itk::FixedArray < double, dimension > EigenValueArrayType;
> typedef itk::Image < EigenValueArrayType, dimension > EigenValueImageType;
> 
> You can get your eigen values then as an image by simply doing a 
> eigenFilter -> GetOutput () 
> call to the eigen filter assigning it to an instance of the above eigen 
> image type.
> 
> Regards,
> Ruben
> 
> 
> Am 24.08.2007 um 15:49 schrieb Patric:
> 
>> Ruben,
>>
>> Do you know how to access the eigenvalue after the 
>> itkSymmetricEigenAnalysisImageF
>> ilter is used?
>>
>> thanks,
>> -Jone
> 
> 
> 
> 
> Hi Karthik,
> 
> 
>> I agree with you on the first point that there is no composite filter 
>> to do all these operations in a memory efficient manner in one filter, 
>> but these needs are so user specific. 
>>
>> As far as a class to compute eigen values and vectors, it does exist 
>> in ITK:
>>
>> See Common/itkSymmetricEigenAnalysis.h
> 
> 
> Sorry, I overlooked, that this is not necessarily an image to image filter.
> 
> 
>> The itkSymmetricEigenAnalysisImageFilter constructs a functor out of 
>> this very class and applies it per pixel. It discards the eigen 
>> vectors .. You could write your own functor, simply by replacing the 
>> following line in itkSymmetricEigenAnalysisImageFilter.h
>>
>>       m_Calculator.ComputeEigenValues( x, eigenValues );
>>
>> with 
>>       m_Calculator.ComputeEigenValuesAndVectors( x, eigenValues, ... );
>>
>> Regards
>> --
>> karthik 
> 
> 
> Regards,
> 
> Ruben
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users