calculate-surface-attributes

Synopsis

calculate-surface-attributes <input> <output> [options]

Description

Calculate attributes of input surface such as normals and curvature. If required, as in case of the curvature calculations, the input mesh is triangulated beforehand if it contains non-triangular faces.

Arguments

input

Input surface mesh.

output

Output surface mesh.

Normals options

-normals

Surface point and cell normals.

-point-normals

Surface point normals.

-cell-normals

Surface cell normals.

-auto-orient, -noauto-orient

Enable/disable auto-orientation of normals. (default: on)

-splitting, -nosplitting

Enable/disable splitting of sharp edges. (default: off)

-consistency, -noconsistency

Enable/disable enforcement of vertex order consistency. (default: on)

-flip-normals, -noflip-normals

Enable/disable flipping of normals. (default: off)

Curvature options

-k1k2 [<name>] [<name>]

Principal curvatures.

-k1 [<name>]

Minimum curvature.

-k2 [<name>]

Maximum curvature.

-e1 [<name>]

Direction of minimum curvature.

-e2 [<name>]

Direction of maximum curvature.

-H [<name>]

Mean curvature: H = .5 * (k1 + k2).

-K [<name>]

Gauss curvature: K = k1 * k2.

-C [<name>]

Curvedness: C = sqrt(.5 * (k1^2 + k2^2)).

-normalize

Normalize curvature using volume of convex hull.

-vtk-curvatures

Use vtkCurvatures when possible.

-robust-curvatures

Do not use vtkCurvatures. Instead, estimate the curvature tensor field and decompose it to obtain principle curvatures. (default)

Parcellation options

-labels <name>

Name of surface point and/or cell parcellation array.

-point-labels <name>

Name of surface point parcellation array.

-cell-labels <name>

Name of surface cell parcellation array.

-border-mask <name>

Add parcellation border mask to output surface mesh, where points/cells adjacent to a given point/cell belong to different parcels have a non-zero value.

Local image options

-image <file>

Input image file required by -gradient* and -patch* options.

-gradient-normal [<name>]

Compute image derivative in normal direction using cubic B-spline interpolation. The <name> of the output point data array is by default ‘ImageGradientNormal’. (default: off)

-gradient-angle [<name>]

Compute cosine of angle made up by image gradient and normal vector using cubic B-spline interpolation for computing the image derivatives. The <name> of the output point data array is by default ‘ImageGradientAngle’ (default: off)

-patch-name <name>

Name of output point data array storing patch image statistics. (default: LocalImageStatistics)

-patch-size <nx> [<ny> [<nz>]]

Size of image patches. When only <nx> is given, an image patch of size nx = ny = nz is used. When only <nz> is omitted, a 2D patch is used.

-patch-spacing <dx> [<dy> [<dz>]]

Spacing between patch sample points. When only <dx> is given, an isotropic sampling in all three dimensions of <dx> is used. When only <dz> is omitted, a 2D patch spacing is used with dz=0.

-patch-space image|world|tangent

Coordinate system of patch. (default: tangent)

• world: Patch is aligned with world coordinate system.
• image: Patch is algined with image coordinate system.
• tangent: Each patch is aligned with the coordinate system made up

by the normal vector and two orthonormal tangent vectors.

-patch-samples, -nopatch-samples

Whether to store individual intensities interpolated at patch sample points.

-demean-patch

Substract mean intensity from individual -patch-samples.

-whiten-patch

Dividide individual -patch-samples by standard deviation.

-patch-min

Append minimum patch intensity to output point data array.

-patch-max

Append maximum patch intensity to output point data array.

-patch-min-abs

Append minimum absolute patch intensity to output point data array.

-patch-max-abs

Append maximum absolute patch intensity to output point data array.

-patch-mean

Append mean patch intensity to output point data array.

-patch-sigma

Append standard deviation of patch intensities to output point data array.

Smoothing options

-smooth-iterations [<niter>]

Number of smoothing iterations.

-smooth-weighting <name> [options]

Smooth scalar attributes using the named weighting function:

• ‘Gaussian’: Isotropic Gaussian smoothing kernel. (default)
• Options: [<sigma>]
• If sigma is not specified, it is automatically determined from the edges.
• ‘AnisotropicGaussian’: Anisotropic Gaussian smoothing kernel.
• Options: [<sigma>] [<sigma2>]
• If sigma is not specified, it is automatically determined from the edges.
• If sigma2 is specified, an anisotropic kernel with standard deviation

sigma along the direction of minimum curvature, and sigma2 in the

direction of maximum curvature is used. - If sigma2 is not specified, an isotropic Gaussian kernel used that is oriented

and scaled along each local geometry axis using the curvature tensor. - ‘InverseDistance’: Inverse node distance. - Options: [<bias>] - If the bias is specified, the distance is estimated as 1/(dist+bias). - ‘Combinatorial’: Uniform node weights.

Standard options

-v, -verbose [n]

Increase/Set verbosity of output messages. (default: 0)

-debug [level]

Increase/Set debug level for output of intermediate results. (default: 0)

-version [major.minor]

Print version and exit or set version to emulate.

-revision

Print revision (or version) number only and exit.

-h, -help

Print help and exit.

Terminal options

-color, -nocolor

Enable/disable colored output. (default: off)

Parallelization options

-threads <n>

Use maximal <n> threads for parallel execution. (default: automatic)