[Insight-users] HessianRecursiveGaussianImageFilter

Thu Jul 5 13:00:05 EDT 2007

Hi Ivan,

Thanks for letting us know about your ITK version,

and thanks for testing the results of the current
implementation.

I'll go ahead and closed both bug reports.

    Thanks

       Luis

-------------------
Iván Macía wrote:
> Hi Luis,
> 
> I am using ITK version 3.2. I have seen this patch is posterior. 
> 
> Unfortunately I think I missed that conversation in the mailing list (I was
> not working those days) and without checking CVS I assumed this had not been
> patched :(
> 
> I checked my results against current CVS and they are exactly the same. The
> only difference is that my patch uses a fixed vector of N
> RecursiveGaussianImageFilters where I change the order in each dimension
> whereas the CVS version uses a std::vector for the zero order filters and
> two members for the gaussian derivatives.
> 
> Thanks for your help
> 
> Ivan
> 
> -----Mensaje original-----
> De: Luis Ibanez [mailto:luis.ibanez at kitware.com] 
> Enviado el: jueves, 05 de julio de 2007 17:47
> Para: Iván Macía
> CC: 'Won-Ki Jeong'; Insight-users at itk.org
> Asunto: Re: [Insight-users] HessianRecursiveGaussianImageFilter
> 
> Hi Ivan,
> 
> Thanks for entering the bug report regarding this filter.
> 
> There seems to be a duplicate report, and an ITK versioning issue
> involved.
> 
> As we just mentioned to Won-Ki:
> 
>     This problem was pointed out by Andreas Keil in the
>     users list on April 2007:
> 
> http://public.kitware.com/pipermail/insight-users/2007-April/022019.html
> 
>     We entered in the Bug tracker as Bug # 4934
> http://public.kitware.com/Bug/bug.php?op=show&bugid=4934
> 
>     and committed a fix in the CVS repository on April 27th 2007:
> http://www.itk.org/cgi-bin/viewcvs.cgi/Code/BasicFilters/itkHessianRecursive
> GaussianImageFilter.txx?root=Insight&sortby=date&r2=1.5&r1=1.4
> 
> 
> 
> 
> What version of ITK did you use for this recent test ?
> 
> Was it a CVS version posterior to April 27th 2007 ?
> 
> 
>    Please let us know,
> 
> 
>       Thanks
> 
> 
> 
> 
> 
> -------------------
> Iván Macía wrote:
> 
>>Hi Won-Ki,
>>
>>Regarding Q1 you are right, the filter computes in the 3D case the six
>>possible gaussian second order derivatives. Note that the filter uses
>>itk::RecursiveGaussianImageFilter internally.
>>
>>Regarding Q2 I think you are also right. When computing second
> 
> orderivative
> 
>>(G*I)xx the filter applies recursive gaussian smoothing with first order
>>derivative twice in the x direction, but the number of smoothing filter is
>>fixed in the static member NumberOfSmoothingFilters = ImageDimension-2, so
>>in the 3D case there is only a single smoothing filter.
>>
>>I made a small test with a 3D MRI image, applied HessianRecursiveGaussian
>>filter with sigma = 5.0 and extracted the six components of the hessian
>>tensor. The attached image shows an slice of the original image and the
>>sixth component (G*I)zz of the hessian matrix of the same slice. It seems
>>like there is smoothing in x but no smoothing in y (see the narrow
>>horizontal lines).
>>
>>I made a Bug entry in the bug tracker with ID 5292
>>http://www.itk.org/Bug/bug.php?op=show&bugid=5292
>>
>>Best regards
>>
>>Ivan
>>
>>
>>-----Mensaje original-----
>>De: insight-users-bounces+imacia=vicomtech.es at itk.org
>>[mailto:insight-users-bounces+imacia=vicomtech.es at itk.org] En nombre de
>>Won-Ki Jeong
>>Enviado el: lunes, 02 de julio de 2007 21:26
>>Para: Insight-users at itk.org
>>Asunto: [Insight-users] HessianRecursiveGaussianImageFilter
>>
>>Hi all
>>
>>I have some questions about the HessianRecursiveGaussianImageFilter.
>>
>>Q1. Is this filter (theoretically) same as applying Gaussian smoothing
>>and compute derivatives to the input image? For example, let G be a
>>Gaussian kernel and I be the input image, then is the filter compute
>>(G*I)xx, (G*I)xy, (G*I)xz, (G*I)yy, (G*I)yz, (G*I)zz (subscript means
>>partial derivative along that axis)?
>>
>>Q2. If so, the implementation is not correct because, for example
>>(G*I)xx, the filter applies 1D recursive Gaussian order-1 derivative
>>filter twice along x and applies 1D recursive Gaussian filter along y,
>>that means along x the Gaussian is applied twice but there is no
>>Gaussian for z direction. Can somebody clarify this?
>>
>>The implementation is in void GenerateData(void) in
>>itkHessianRecursiveGaussianImageFilter.txx.
>>
>>Thanks a lot!
>>
>>Won-Ki
>>_______________________________________________
>>Insight-users mailing list
>>Insight-users at itk.org
>>http://www.itk.org/mailman/listinfo/insight-users
>>
>>No virus found in this incoming message.
>>Checked by AVG Free Edition. 
>>Version: 7.5.476 / Virus Database: 269.9.14/884 - Release Date: 02/07/2007
>>15:35
>> 
>>
>>No virus found in this outgoing message.
>>Checked by AVG Free Edition. 
>>Version: 7.5.476 / Virus Database: 269.9.14/885 - Release Date: 03/07/2007
>>10:02
>> 
>>
>>_______________________________________________
>>Insight-users mailing list
>>Insight-users at itk.org
>>http://www.itk.org/mailman/listinfo/insight-users
>>
> 
> 
> No virus found in this incoming message.
> Checked by AVG Free Edition. 
> Version: 7.5.476 / Virus Database: 269.10.0/886 - Release Date: 04/07/2007
> 13:40
>  
> 
> No virus found in this outgoing message.
> Checked by AVG Free Edition. 
> Version: 7.5.476 / Virus Database: 269.10.0/886 - Release Date: 04/07/2007
> 13:40
>  
> 
> 