[Insight-users] ApproximateSignedDistanceMapImageFilter requires binary image ??

[Ricardo A Corredor J]

Hi,

I'm using this class in order to generate the distance map of a binary
image by now. However, I did a naive test with an unsigned short image and
surprise! the filter worked and very well actually, setting min and max
unsigned-short values. Looking deep in the code I see
that IsoContourDistanceImageFilter doesn't have the restriction of binary
images and it works almost as a level-set segmentation step with a specific
iso-value defined by inside and outside value (min and max in this case).

I'm concerned about the results of this filter and their reliability in
different images. Krissian's article don't talk to much about that and I
would like to know which could be a cause of bad results using it and if
there are specific reasons of imposing a binary-image restriction. Is it
crazy to use it in non-binary images?

Thanks in advance!

Saludos,


-------------------------------------------------------------
*Ricardo A. Corredor*
*RaC*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20121125/544fef16/attachment.htm>