[MINC-users] Filling with mincreshape

Simon Fristed Eskildsen se at hst.aau.dk
Tue Jun 19 11:49:55 EDT 2007


Hi all,

I never resolved the issue with mincreshape, and now I have stumbled 
upon a similar problem. If I do

mincreshape -rowsize 256 -colsize 256 test.mnc test_reshaped.mnc

and test.mnc is signed short (and standard mni space size) mincreshape 
fills the "new" voxels with what seems to be the maximum value in that 
particular slice. However, if I do

mincreshape -unsigned test.mnc test_usign.mnc
mincreshape -rowsize 256 -colsize 256 test_usign.mnc test_reshaped.mnc

everything seems ok. But if I look closely, also this volume has 
non-zero filling. Three individual slices are uniformly filled.

It seems that mincreshape uses uninitialized mem.

Simon

Simon Fristed Eskildsen wrote:
> Andrew Janke wrote:
>> First, the -fill and -fillvalue options are mutually exclusive of each
>> other. Meaning choose one of -nofill -fill and -fillvalue.  in this
>> case, I think you want to skip the -fill option.
>>
>> Try this again without the -fill option.  Also check out -pixfillvalue
>> it might be closer to what you want.
>>
> 
> I think I have tried all combinations of -fill, -fillvalue, -pixfill, 
> and -pixfillvalue all with the same result. A rudimentary investigation 
> of the source code reveals that mincreshape decides that there is no 
> need to fill anything. Specifically, the variable 
> reshape_info->need_fillvalue is set to false. Maybe I am doing the 
> extrapolation wrong?
> 
> Simon
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users


More information about the MINC-users mailing list