[MINC-users] Registration of Binary Images

[Mallar Chakravarty]

Hi Elizabeth,

Another neat trick that I find often works is to manufacture some contrast
be estimating a distance map using either mincchamfer or mincmorph.

mallar


On Thu, Jun 12, 2014 at 4:34 PM, Andrew Janke <a.janke at gmail.com> wrote:

> Hi Elizabeth,
>
> Have you tried to blur them with a FWHM kernel of about 5mm before
> registration?
>
>
> a
>
>
> On 13 June 2014 04:51, Elizabeth Wicks <elizabethw at msmri.medicine.ubc.ca>
> wrote:
> > Hi all,
> >
> > I was wondering if minctracc provides any options to help optimize
> > registration of two binary images? In particular, the registration of two
> > skull images (with pixel intensities of either 1000 or 0) . Currently,
> there
> > doesn't seem to be enough information in the images to do a consistent
> > reliable -lsq12 minctracc registration.
> >
> > Thank you in advance!
> >
> > Elizabeth
> > _______________________________________________
> > MINC-users at bic.mni.mcgill.ca
> > http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>