[MINC-users] ray_trace colour maps and rendering labels

[EJ Nikelski]

Hello minc-users,

   Ok, for posterity and for all future Googlers who find their way to
this post, here is what I've figured out about how to create one's own
*discrete* ray_trace-friendly color map -- useful for rendering
labels.  Pretty simple, when you know how ;)

(1) file type is ".ccd", so use this as an extension

(2) the first column must contain the label number, which *must* be
monotonic. That is, keep your labels in ascending order.

(3) subsequent columns are used to define the corresponding color.
This can be done in the usual way: e.g., [ r g b ], [ r g b a ], or [
blue ].  All fields are space-delimited.

(4) each label in the volume being rendered *should* have a
corresponding line in the .ccd file, else ray_trace might make
something up.

(5) unused label values do *not* need a line defined in the .ccd file.
So, if you have a .ccd range of 128, but only need 10 label values,
you need only define the 10 labels that you need.

(6) the ray_trace <min> and <max> values specified on the "-usercc"
argument, *must* have corresponding labels in the .ccd file, else
labels get unexpectedly assigned. So, if in ray_trace you say ...
>>-usercc    user_map.ccd  0 127  volume.mnc  -1 1.0

   ... then your .ccd file better have label entries for "0" and "127".

   That's all that I can think of for now.  If anyone has any other
suggestions, pls feel free to contribute.  Now get out there and spend
your weekend creating label-friendly color maps ... I will <sigh>.

Cheers,

-Jim



On Thu, Mar 25, 2010 at 7:47 PM, Claude LEPAGE <claude at bic.mni.mcgill.ca> wrote:
> Jim,
>
>>   Can I add one additional follow-up?  First, for context, I've fit
>> the AAL labels (manual labeling of the colin-27 brain) to the Civet
>> model (fit to a 4-mm grid), and am now applying my subject's inverted
>> non-linear transform to the AAL volume in order to fit the labels onto
>> my subject's linear fit T1 (out of Civet). The results look good, but
>> ... the "real" values in my label volume appear to be stored as floats
>> -- or so says register, Display and postf.  How do I convert my label
>> volume into simply having "real" values =3D voxel values =3D label values
>> =3D integers?
>
> Try something like:
>  mincreshape -byte -image_range 0 255 -valid_range 0 255 ...
>
> Claude
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>



-- 
=================================
Jim Nikelski, Ph.D.
Postdoctoral Research Fellow
Bloomfield Centre for Research in Aging
Lady Davis Institute for Medical Research
Sir Mortimer B. Davis - Jewish General Hospital
McGill University