Usage: 3dAutobox [options] DATASET
Computes size of a box that fits around the volume.
Also can be used to crop the volume to that box.

-prefix PREFIX = Crop the input dataset to the size of the box, and
                 write an output dataset with PREFIX for the name.
               * If -prefix is not used, no new volume is written out,
                 just the (x,y,z) extents of the voxels to be kept.

-input DATASET = An alternate way to specify the input dataset.
                 The default method is to pass DATASET as
                 the last parameter on the command line.

-noclust       = Don't do any clustering to find box. Any non-zero
                 voxel will be preserved in the cropped volume.
                 The default method uses some clustering to find the
                 cropping box, and will clip off small isolated blobs.

-extent: Write to standard out the spatial extent of the box

-extent_ijk    = Write out the 6 auto bbox ijk slice numbers to
                     imin imax jmin jmax kmin kmax
                 Note that resampling would affect the ijk vals (but
                 not necessarily the xyz ones).
                 Also note that this value is calculated before
                 any '-npad ...' option, so it would ignore that.

-extent_ijk_to_file FF = Write out the 6 auto bbox ijk slice numbers to
                 a simple-formatted text file FF (single row file):
                     imin imax jmin jmax kmin kmax
                 (same notes as above apply).

-extent_ijk_midslice = Write out the 3 ijk midslices of the autobox to
                 the screen:
                     imid jmid kmid
                 These are obtained via: (imin + imax)/2, etc.

-extent_xyz_midslice = Write out the 3 xyz midslices of the autobox to
                 the screen:
                     xmid ymid zmid
                 These are obtained via: (xmin + xmax)/2, etc.
                 These follow the same meaning as '-extent'.

-npad NNN      = Number of extra voxels to pad on each side of box,
                 since some troublesome people (that's you, LRF) want
                 this feature for no apparent reason.
               * With this option, it is possible to get a dataset that
                 is actually bigger than the input.
               * You can input a negative value for NNN, which will
                 crop the dataset even more than the automatic method.

++ Compile date = Sep 17 2020 {AFNI_20.2.18:linux_ubuntu_16_64}