[Camino-users] sflutgen problem, SH model

[Ian Malone]

Hi,

I'm trying to use a QBall SH reconstruction with Pico, but am hitting a 
problem at the sflutgen step. Not sure what's causing this, SH order is 
too low maybe? The scheme file has 64 unique DEO, steps were:

qballmx -schemefile DWI.scheme2 -basistype sh -order 6 -normalize > 
qballMatrix_SH6.Bdouble

sfpicocalibdata -schemefile DWI.scheme2 -infooutputfile A.info > 
PICo_data.Bfloat

linrecon PICo_data.Bfloat DWI.scheme2 qballMatrix_SH6.Bdouble -normalize 
 > calib_ODFs_SH6.Bdouble

sfpeaks -inputmodel sh -order 6 -numpds 3 < calib_ODFs_SH6.Bdouble > 
calib_ODFs_SH6_PDs.Bdouble
- complains about not having the scheme file.

sflutgen -infofile A.info -outputstem SH_lut < 
calib_ODFs_SH6_PDs.Bdouble -pdf   bingham

Nov 14, 2011 1:01:46 PM data.ExternalDataSource initFileInput
INFO: reading data from standard input
Nov 14, 2011 1:02:00 PM misc.LoggedException logException
SEVERE: Exception in thread "main" class misc.LoggedException: not 
enough populated bins to enable surface fit.  Number of bins: 3
        at apps.GenerateMFR_LUT.fitSurface(GenerateMFR_LUT.java:570)
        at apps.GenerateMFR_LUT.main(GenerateMFR_LUT.java:823)

Exception in thread "main" misc.LoggedException: not enough populated 
bins to enable surface fit.  Number of bins: 3
        at apps.GenerateMFR_LUT.fitSurface(GenerateMFR_LUT.java:570)
        at apps.GenerateMFR_LUT.main(GenerateMFR_LUT.java:823)
malone at img-129:~/git/conmap-framework.git/altscheme

Thanks for your time,
Ian