[Insight-users] GradientAnistropicDiffusionImageFilter

Joshua Cates cates at sci . utah . edu
Wed, 21 May 2003 09:52:20 -0600 (MDT)


Hi Samuel,

The GradientAnisotropicDiffusionFilter is essentially classic Perona-Malik
anisotropic diffusion.  The only differences are the extension to ND and a
better gradient approximation over that described in their original paper.

Josh.

______________________________
 Josh Cates			
 School of Computer Science	
 University of Utah
 Email: cates@sci.utah.edu
 Phone: (801) 587-7697
 URL:   http://www.sci.utah.edu/~cates


On Wed, 21 May 2003, Samuel Rodríguez Bescos wrote:

> Hi everybody!!!!
> 
> I have a question related to the anisotropic diffusion. In the
> itkSoftwareGuide It's said that the Smoothed Image is the solution to
> the heat equation. With g(x,y,t) the result of the convolution of the
> image with a Gaussian:
> 
> g(x,y,t) = G ( sqrt(2t)) * f (x,y )
> 
> with f (x,y) the original image. (pages 91 and 92) 
> 
> 
> is this fact related to Catté revision of the method of Perona and Malik
> in order to reduce the noise effects of the image?
> 
> Thanks in advance
> 
> Sam
>