[Insight-users] Distance map computation on an empty image

[Juan Cardelino]

In my case, I think this problem is also related to the depth of the
image. With a zero input image you get that ramp, which is wrong. But,
if I use the input image with 8 bit depth I get  the same, probably
because the threshold to determine the input points is fixed and not
relative to the depth of the image.
Regards,
              Juan

>
> I got the same result even with a non-empty image, maybe my problem is
> that the pixels I think there are white, and below some threshold and
> taken as zero. But anyway, I've used the same method to generate a
> couple of input masks, same images size but different contents. In one
> case it works and in the other gives the same ramp as yours.
>
> working example:
> SignedDanielssonDistanceMapImageFilter example.mask.01.png map.01.mha
> a.mha b.mha
> non-working example:
> SignedDanielssonDistanceMapImageFilter example.mask.02.png map.02.mha
> a.mha b.mha
>
> you can find both input and output images at
> http://iie.fing.edu.uy/~juanc/files/tmp
>
> Seems a bug to me too.
> Regards,
>                 Juan
>