[Insight-users] slice based memory representation of itk volume

[Kevin H. Hobbs]

On Thu, 2005-11-03 at 14:04 -0800, Feng Ma wrote:
> Kevin and James:
> 
>   Thanks a lot for the information.
> 
>   I am actually looking at a filter that could do Gaussian smoothing like 
> itkDiscreteGaussianImageFilter, without using a continous 1D block of 
> memory.
> 
>   To use existing itk filters, probably the best shot is to use sparse 
> image. I think itkDiscreteGaussianImageFilter does not support sparse image. 
> Does the itkSmoothingRecursiveGaussianImageFilter support sparse image? Or 
> there is any other filter that I could use/tweak to achieve my goal?
> 
>   Thanks.
> 
> -Feng
> 

You can always fake it.  I get around the memory troubles I have when I
load all slices of an image at once, by loading only a few slices at a
time.  You could always just load the slice you want to filter and the
slices that would fill the neighborhood in the z ( or n for n-d ).  If
you share your technique I'm sure others will be interested and will
help develop it.  It'd be like streaming the image, but more work.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://public.kitware.com/pipermail/insight-users/attachments/20051104/47aa5269/attachment-0001.pgp