minc_nuyl

Intensity normalization using piece-wise linear histogram matching.

minc_nuyl [options] <source.mnc> <target.mnc> [output]

DESCRIPTION

minc_nuyl performs intensity normalization of a source MINC volume to match the intensity histogram of a target volume using the piece-wise linear histogram standardization method. The method is based on Nyul, Udupa, and Zhang (2000), “New Variants of a Method of MRI Scale Standardization” (IEEE Transactions on Medical Imaging, 19(2):143-150).

The tool can also create a standardized histogram model from a single volume using the --chist option, and apply normalization using a look-up table.

OPTIONS

--source-mask mask

Apply a mask to the source volume for histogram computation.

--target-mask mask

Apply a mask to the target volume for histogram computation.

--linear

Use global linear normalization instead of piece-wise linear mapping.

--steps n

Set the number of histogram matching steps (landmarks). Default: 10.

--fix_zero_padding

Handle zero-padded regions in the input volumes.

--verbose

Print progress information during processing.

--ks

Report the Kolmogorov-Smirnov statistic between the normalized and target histograms.

--chist

Create a standardized histogram file from the source volume instead of normalizing. In this mode, the second argument is the output histogram filename.

--dump lut

Write the intensity mapping look-up table to the specified file.

--cut-off pct

Set the percentile cut-off for histogram matching.

EXAMPLES

Normalize source to match target intensity:

minc_nuyl source.mnc target.mnc normalized.mnc

Normalize with brain masks:

minc_nuyl --source-mask src_mask.mnc --target-mask tgt_mask.mnc \
    source.mnc target.mnc normalized.mnc

Create a standardized histogram:

minc_nuyl source.mnc model.hist --chist

Normalize with linear mapping:

minc_nuyl --linear source.mnc target.mnc normalized.mnc

AUTHOR

Vladimir S. Fonov - McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University.

COPYRIGHTS

Copyright © 2011 by Vladimir S. Fonov

SEE ALSO

volume_pol, inormalize