mincchamfer

compute the chamfer distance transform of a MINC volume

mincchamfer [<options>] <inputfile> <outputfile>

DESCRIPTION

mincchamfer reads a 3D MINC volume and computes the chamfer distance transform, writing the result to an output MINC volume. The chamfer distance is an approximation of the Euclidean distance transform: for each voxel, it estimates the distance to the nearest non-background (boundary) voxel using a multi-pass local neighbourhood propagation algorithm.

The maximum distance value can be capped via the -max_dist option. This is useful in registration pipelines for computing distance maps from segmented structures or binary masks, which can then be used as features for registration cost functions.

OPTIONS

-max_dist value

Maximum distance value in the output transform. Default: 50.

-first n

Number of initial background structure dilations. Default: 0.

-verbose

Write messages indicating progress (default).

-quiet

Do not write log messages.

-debug

Print out debugging information.

-help

Print summary of command-line options and exit.

-version

Print the program’s version number and exit.

EXAMPLES

Compute the chamfer distance transform of a binary mask:

mincchamfer mask.mnc distance_map.mnc

Compute the distance transform with a maximum distance of 100mm:

mincchamfer -max_dist 100 mask.mnc distance_map.mnc

AUTHOR

Louis Collins - McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University.

COPYRIGHTS

Copyright © 1993 by Louis Collins

SEE ALSO

minctracc mincmorph dilate_volume