[Insight-users] Warping w.o. overall intensity intergral change
Torsten Rohlfing
torsten at synapse.sri.com
Thu May 28 12:25:51 EDT 2009
Hi Felix.
The intensity integral is conceptually a product of intensity times
volume (area in 2D, length in 1D). So if you increase the volume of
high-intensity regions due to warping, the overall intensity integral
increases as a result.
The following are your options:
1. globally normalize the intensity integral, i.e., multiply the
intensity of each pixel in the warped image with IntOriginal/IntWarped,
where "IntOriginal" is the original integral and "IntWarped" is the
integral after warping (but before normalization). This will guarantee a
globally invariant intensity integral, but may be inappropriate for your
application.
2. locally correct for volume changes by multiplying warped intensities
with the Jacobian of the deformation at that pixel. This is physically
the correct way to do things in some applications where signal piles up
due to geometrical distortion, for example in echo-planar MR imaging.
Hope this helps.
Torsten
> Hello ITK Users,
>
> I am warping a functional measurment 3D volume using a spline grid for
> pixel displacement, as discribed in the Softwareguide.
>
> The quality of the warped volume is fine by visual inspection is fine,
> however,
> the overall intensity integral increase by x1.25.
>
>
> For analysis in need the intensity integral invariant of course, so:
>
> (0. why does the warping increase overall intensity)
>
> 1. Is there a way to maintain overall intensity values in warping
> - i.e. gaussian interp. kernel etc. ?
>
>
>
--
Torsten Rohlfing, PhD SRI International, Neuroscience Program
Senior Research Scientist 333 Ravenswood Ave, Menlo Park, CA 94025
Phone: ++1 (650) 859-3379 Fax: ++1 (650) 859-2743
torsten at synapse.sri.com http://www.stanford.edu/~rohlfing/
"Though this be madness, yet there is a method in't"
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: GnuPG v1.4.9 (GNU/Linux)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=masI
-----END PGP PUBLIC KEY BLOCK-----
More information about the Insight-users
mailing list