[Insight-users] Computing eigenvectors and eigenvalues

[Miller, James V (Research)]

I would think the vnl_real_eigensystem should fit your needs. You 
could use the vnl_generalized_eigensystem but that solves a more 
complicated eigensystem problem ( A*x = lambda*B*x).

Note that the JacobiN routine in VTK is only suitable for real
symmetric matrics.  So if your matrix is not symmetric, you wouldn't
be able to compute the eigensystem in VTK using the Jacobi interation
routines.



> -----Original Message-----
> From: Adolfo Gonzalez Uzabal [mailto:fito at unizar . es]
> Sent: Wednesday, September 17, 2003 10:09 AM
> To: insight-users at itk . org
> Subject: [Insight-users] Computing eigenvectors and eigenvalues
> 
> 
> 
> Hello everybody!
> 
> I have to compute eigenvectors and eigenvalues of a matrix.
> 
> I'm searching for something like vtkMath::JacobiN routine, 
> but I'm not able of 
> finding it.
> 
> I know about vnl_complex, vnl_generalized, vnl_symmetric and 
> vnl_sparse_symmetric_system routines, but I find those 
> unappropriate for one 
> thing or the other (my matrix is not symmetric, nor sparse, 
> nor complex, and so 
> on). 
> 
> I'm thinking "vnl_real_eigensystem" fits my trouble but I'm 
> not sure about the 
> results I'm getting.
> 
> Did anyone have any clue?
> 
> Sincerely yours.
> Adolfo G.U.
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk . org
> http://www . itk . org/mailman/listinfo/insight-users
>