DOKK / manpages / debian 11 / mia-tools / mia-3dnonrigidreg.1.en
mia-3dnonrigidreg(1) General Commands Manual mia-3dnonrigidreg(1)

mia-3dnonrigidreg - Non-linear registration of 3D images

mia-3dnonrigidreg -i <in-image> -r <ref-image> -o <out-image> [options] <PLUGINS:3dimage/fullcost>

mia-3dnonrigidreg This program implements the registration of two gray scale 3D images.

verbosity of output, print messages of given level and higher priorities. Supported priorities starting at lowest level are:

trace ‐ Function call trace
debug ‐ Debug output
info ‐ Low level messages
message ‐ Normal messages
warning ‐ Warnings
fail ‐ Report test failures
error ‐ Report errors
fatal ‐ Report only fatal errors
print copyright information

print this help

-? --usage
print a short help

print the version number and exit

test image
For supported file types see PLUGINS:3dimage/io
reference image
For supported file types see PLUGINS:3dimage/io
registered output image
For supported file types see PLUGINS:3dimage/io
output transformation
For supported file types see PLUGINS:3dtransform/io

Maxiumum number of threads to use for processing,This number should be lower or equal to the number of logical processor cores in the machine. (-1: automatic estimation).

multi-resolution levels

Optimizer used for minimization
For supported plugins see PLUGINS:minimizer/singlecost
transformation type
For supported plugins see PLUGINS:3dimage/transform

Central difference filter kernel, mirror boundary conditions are used.

(no parameters)
spacial Gauss filter kernel, supported parameters are:

w = 1; uint in [0, inf)
half filter width.

This plugin provides the 1D folding kernel for the Scharr gradient filter

(no parameters)

Spline interpolation boundary conditions that mirror on the boundary

(no parameters)
Spline interpolation boundary conditions that repeats the value at the boundary

(no parameters)
Spline interpolation boundary conditions that assumes zero for values outside

(no parameters)

B-spline kernel creation , supported parameters are:

d = 3; int in [0, 5]
Spline degree.

OMoms-spline kernel creation, supported parameters are:

d = 3; int in [3, 3]
Spline degree.

Image combiner 'absdiff'

(no parameters)
Image combiner 'add'

(no parameters)
Image combiner 'div'

(no parameters)
Image combiner 'mul'

(no parameters)
Image combiner 'sub'

(no parameters)

local normalized cross correlation with masking support., supported parameters are:

w = 5; uint in [1, 256]
half width of the window used for evaluating the localized cross correlation.

Spline parzen based mutual information., supported parameters are:

cut = 0; float in [0, 40]
Percentage of pixels to cut at high and low intensities to remove outliers.

mbins = 64; uint in [1, 256]
Number of histogram bins used for the moving image.

mkernel = [bspline:d=3]; factory
Spline kernel for moving image parzen hinstogram. For supported plug-ins see PLUGINS:1d/splinekernel

rbins = 64; uint in [1, 256]
Number of histogram bins used for the reference image.

rkernel = [bspline:d=0]; factory
Spline kernel for reference image parzen hinstogram. For supported plug-ins see PLUGINS:1d/splinekernel

normalized cross correlation.

(no parameters)
This function evaluates the image similarity based on normalized gradient fields. Given normalized gradient fields $ _S$ of the src image and $ _R$ of the ref image various evaluators are implemented., supported parameters are:

eval = ds; dict
plugin subtype (sq, ds,dot,cross). Supported values are:
ds ‐ square of scaled difference
dot ‐ scalar product kernel
cross ‐ cross product kernel

3D image cost: sum of squared differences, supported parameters are:

autothresh = 0; float in [0, 1000]
Use automatic masking of the moving image by only takeing intensity values into accound that are larger than the given threshold.

norm = 0; bool
Set whether the metric should be normalized by the number of image pixels.

3D image cost: sum of squared differences, with automasking based on given thresholds, supported parameters are:

rthresh = 0; double
Threshold intensity value for reference image.

sthresh = 0; double
Threshold intensity value for source image.

intensity bandpass filter, supported parameters are:

max = 3.40282e+38; float
maximum of the band.

min = 0; float
minimum of the band.

image binarize filter, supported parameters are:

max = 3.40282e+38; float
maximum of accepted range.

min = 0; float
minimum of accepted range.

morphological close, supported parameters are:

hint = black; string
a hint at the main image content (black|white).

shape = [sphere:r=2]; factory
structuring element. For supported plug-ins see PLUGINS:3dimage/shape

Combine two images with the given combiner operator. if 'reverse' is set to false, the first operator is the image passed through the filter pipeline, and the second image is loaded from the file given with the 'image' parameter the moment the filter is run., supported parameters are:

image =(required, input, io)
second image that is needed in the combiner. For supported file types see PLUGINS:3dimage/io

op =(required, factory)
Image combiner to be applied to the images. For supported plug-ins see PLUGINS:3dimage/combiner

reverse = 0; bool
reverse the order in which the images passed to the combiner.

image pixel format conversion filter, supported parameters are:

a = 1; float
linear conversion parameter a.

b = 0; float
linear conversion parameter b.

map = opt; dict
conversion mapping. Supported values are:
copy ‐ copy data when converting
linear ‐ apply linear transformation x -> a*x+b
range ‐ apply linear transformation that maps the input data type range to the output data type range
opt ‐ apply a linear transformation that maps the real input range to the full output range
optstat ‐ apply a linear transform that maps based on input mean and variation to the full output range

repn = ubyte; dict
output pixel type. Supported values are:
bit ‐ binary data
sbyte ‐ signed 8 bit
ubyte ‐ unsigned 8 bit
sshort ‐ signed 16 bit
ushort ‐ unsigned 16 bit
sint ‐ signed 32 bit
uint ‐ unsigned 32 bit
slong ‐ signed 64 bit
ulong ‐ unsigned 64 bit
float ‐ floating point 32 bit
double ‐ floating point 64 bit
none ‐ no pixel type defined

Crop a region of an image, the region is always clamped to the original image size in the sense that the given range is kept., supported parameters are:

end = [[4294967295,4294967295,4294967295]]; streamable
end of cropping range, maximum = (-1,-1,-1).

start = [[0,0,0]]; streamable
begin of cropping range.

3d image stack dilate filter, supported parameters are:

hint = black; string
a hint at the main image content (black|white).

shape = [sphere:r=2]; factory
structuring element. For supported plug-ins see PLUGINS:3dimage/shape

Evaluate the 3D distance transform of an image. If the image is a binary mask, then result of the distance transform in each point corresponds to the Euclidian distance to the mask. If the input image is of a scalar pixel value, then the this scalar is interpreted as heighfield and the per pixel value adds to the distance.

(no parameters)
Downscale the input image by using a given block size to define the downscale factor. Prior to scaling the image is filtered by a smoothing filter to eliminate high frequency data and avoid aliasing artifacts., supported parameters are:

b = [[1,1,1]]; 3dbounds
blocksize.

bx = 1; uint in [1, inf)
blocksize in x direction.

by = 1; uint in [1, inf)
blocksize in y direction.

bz = 1; uint in [1, inf)
blocksize in z direction.

kernel = gauss; factory
smoothing filter kernel to be applied, the size of the filter is estimated based on the blocksize.. For supported plug-ins see PLUGINS:1d/spacialkernel

3d image stack erode filter, supported parameters are:

hint = black; string
a hint at the main image content (black|white).

shape = [sphere:r=2]; factory
structuring element. For supported plug-ins see PLUGINS:3dimage/shape

isotropic 3D gauss filter, supported parameters are:

w = 1; int in [0, inf)
filter width parameter.

3D image to gradient norm filter

(no parameters)
Use an input binary mask and a reference gray scale image to do region growing by adding the neighborhood pixels of an already added pixel if the have a lower intensity that is above the given threshold., supported parameters are:

min = 1; float
lower threshold for mask growing.

ref =(required, input, io)
reference image for mask region growing. For supported file types see PLUGINS:3dimage/io

shape = 6n; factory
neighborhood mask. For supported plug-ins see PLUGINS:3dimage/shape

intensity invert filter

(no parameters)
This filter scales an image to make the voxel size isometric and its size to correspond to the given value, supported parameters are:

interp = [bspline:d=3]; factory
interpolation kernel to be used . For supported plug-ins see PLUGINS:1d/splinekernel

size = 1; float in (0, inf)
isometric target voxel size.

3D image k-means filter. In the output image the pixel value represents the class membership and the class centers are stored as attribute in the image., supported parameters are:

c = 3; int in [2, inf)
number of classes.

A filter to label the connected components of a binary image., supported parameters are:

n = 6n; factory
neighborhood mask. For supported plug-ins see PLUGINS:3dimage/shape

Image filter to remap label id's. Only applicable to images with integer valued intensities/labels., supported parameters are:

map =(required, input, string)
Label mapping file.

A filter that only creates output voxels that are already created in the input image. Scaling is done by using a voting algorithms that selects the target pixel value based on the highest pixel count of a certain label in the corresponding source region. If the region comprises two labels with the same count, the one with the lower number wins., supported parameters are:

out-size =(required, 3dbounds)
target size given as two coma separated values.

Load the input image from a file and use it to replace the current image in the pipeline., supported parameters are:

file =(required, input, io)
name of the input file to load from.. For supported file types see PLUGINS:3dimage/io

This is a label voting downscale filter. It adownscales a 3D image by blocks. For each block the (non-zero) label that appears most times in the block is issued as output pixel in the target image. If two labels appear the same number of times, the one with the lower absolute value wins., supported parameters are:

b = [[1,1,1]]; 3dbounds
blocksize for the downscaling. Each block will be represented by one pixel in the target image..

Mask an image, one image is taken from the parameters list and the other from the normal filter input. Both images must be of the same dimensions and one must be binary. The attributes of the image coming through the filter pipeline are preserved. The output pixel type corresponds to the input image that is not binary., supported parameters are:

input =(required, input, io)
second input image file name. For supported file types see PLUGINS:3dimage/io

3D image mean filter, supported parameters are:

w = 1; int in [1, inf)
half filter width.

median 3d filter, supported parameters are:

w = 1; int in [1, inf)
filter width parameter.

Mean of Least Variance 3D image filter, supported parameters are:

w = 1; int in [1, inf)
filter width parameter.

3D image mean-sigma normalizing filter, supported parameters are:

w = 1; int in [1, inf)
half filter width.

morphological open, supported parameters are:

hint = black; string
a hint at the main image content (black|white).

shape = [sphere:r=2]; factory
structuring element. For supported plug-ins see PLUGINS:3dimage/shape

3D image reorientation filter, supported parameters are:

map = xyz; dict
oriantation mapping to be applied. Supported values are:
xyz ‐ keep orientation
p-yzx ‐ permutate x->z->y->x
p-zxy ‐ permutate x->y->z->x
f-yz ‐ flip y-z
f-xy ‐ flip x-y
f-xz ‐ flip x-z
r-x90 ‐ rotate around x-axis clockwise 90 degree
r-x180 ‐ rotate around x-axis clockwise 180 degree
r-x270 ‐ rotate around x-axis clockwise 270 degree
r-y90 ‐ rotate around y-axis clockwise 90 degree
r-y180 ‐ rotate around y-axis clockwise 180 degree
r-y270 ‐ rotate around y-axis clockwise 270 degree
r-z90 ‐ rotate around z-axis clockwise 90 degree
r-z180 ‐ rotate around z-axis clockwise 180 degree
r-z270 ‐ rotate around z-axis clockwise 270 degree

Resize an image. The original data is centered within the new sized image., supported parameters are:

size = [[0,0,0]]; streamable
new size of the image a size 0 indicates to keep the size for the corresponding dimension..

salt and pepper 3d filter, supported parameters are:

thresh = 100; float in [0, inf)
thresh value.

w = 1; int in [1, inf)
filter width parameter.

3D image filter that scales to a given target size , supported parameters are:

interp = [bspline:d=3]; factory
interpolation kernel to be used . For supported plug-ins see PLUGINS:1d/splinekernel

s = [[0,0,0]]; 3dbounds
target size to set all components at once (component 0:use input image size).

sx = 0; uint in [0, inf)
target size in x direction (0:use input image size).

sy = 0; uint in [0, inf)
target size in y direction (0:use input image size).

sz = 0; uint in [0, inf)
target size in y direction (0:use input image size).

The 3D Scharr filter for gradient evaluation. Note that the output pixel type of the filtered image is the same as the input pixel type, so converting the input beforehand to a floating point valued image is recommendable., supported parameters are:

dir = x; dict
Gradient direction. Supported values are:
x ‐ gradient in x-direction
y ‐ gradient in y-direction
z ‐ gradient in z-direction

A filter that creats a binary mask representing the intensity with the highest pixel count.The pixel value 0 will be ignored, and if two intensities have the same pixel count, then the result is undefined. The input pixel must have an integral pixel type.

(no parameters)
3D image intensity separaple convolution filter, supported parameters are:

kx = [gauss:w=1]; factory
filter kernel in x-direction. For supported plug-ins see PLUGINS:1d/spacialkernel

ky = [gauss:w=1]; factory
filter kernel in y-direction. For supported plug-ins see PLUGINS:1d/spacialkernel

kz = [gauss:w=1]; factory
filter kernel in z-direction. For supported plug-ins see PLUGINS:1d/spacialkernel

The 2D Sobel filter for gradient evaluation. Note that the output pixel type of the filtered image is the same as the input pixel type, so converting the input beforehand to a floating point valued image is recommendable., supported parameters are:

dir = x; dict
Gradient direction. Supported values are:
x ‐ gradient in x-direction
y ‐ gradient in y-direction
z ‐ gradient in z-direction

seeded watershead. The algorithm extracts exactly so many reagions as initial labels are given in the seed image., supported parameters are:

grad = 0; bool
Interpret the input image as gradient. .

mark = 0; bool
Mark the segmented watersheds with a special gray scale value.

n = [sphere:r=1]; factory
Neighborhood for watershead region growing. For supported plug-ins see PLUGINS:3dimage/shape

seed =(required, input, string)
seed input image containing the lables for the initial regions.

Save the input image to a file and also pass it through to the next filter, supported parameters are:

file =(required, output, io)
name of the output file to save the image too.. For supported file types see PLUGINS:3dimage/io

3D morphological thinning, based on: Lee and Kashyap, 'Building Skeleton Models via 3-D Medial Surface/Axis Thinning Algorithms', Graphical Models and Image Processing, 56(6):462-478, 1994. This implementation only supports the 26 neighbourhood.

(no parameters)
Transform the input image with the given transformation., supported parameters are:

file =(required, input, io)
Name of the file containing the transformation.. For supported file types see PLUGINS:3dtransform/io

imgboundary = ; factory
override image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = ; factory
override image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

3D image variance filter, supported parameters are:

w = 1; int in [1, inf)
half filter width.

basic watershead segmentation., supported parameters are:

evalgrad = 0; bool
Set to 1 if the input image does not represent a gradient norm image.

mark = 0; bool
Mark the segmented watersheds with a special gray scale value.

n = [sphere:r=1]; factory
Neighborhood for watershead region growing. For supported plug-ins see PLUGINS:3dimage/shape

thresh = 0; float in [0, 1)
Relative gradient norm threshold. The actual value threshold value is thresh * (max_grad - min_grad) + min_grad. Bassins separated by gradients with a lower norm will be joined.

Generalized image similarity cost function that also handles multi-resolution processing. The actual similarity measure is given es extra parameter., supported parameters are:

cost = ssd; factory
Cost function kernel. For supported plug-ins see PLUGINS:3dimage/cost

debug = 0; bool
Save intermediate resuts for debugging.

ref =(input, io)
Reference image. For supported file types see PLUGINS:3dimage/io

src =(input, io)
Study image. For supported file types see PLUGINS:3dimage/io

weight = 1; float
weight of cost function.

Similarity cost function that maps labels of two images and handles label-preserving multi-resolution processing., supported parameters are:

maxlabel = 256; int in [2, 32000]
maximum number of labels to consider.

ref =(input, io)
Reference image. For supported file types see PLUGINS:3dimage/io

src =(input, io)
Study image. For supported file types see PLUGINS:3dimage/io

weight = 1; float
weight of cost function.

Generalized masked image similarity cost function that also handles multi-resolution processing. The provided masks should be densly filled regions in multi-resolution procesing because otherwise the mask information may get lost when downscaling the image. The mask may be pre-filtered - after pre-filtering the masks must be of bit-type.The reference mask and the transformed mask of the study image are combined by binary AND. The actual similarity measure is given es extra parameter., supported parameters are:

cost = ssd; factory
Cost function kernel. For supported plug-ins see PLUGINS:3dimage/maskedcost

ref =(input, io)
Reference image. For supported file types see PLUGINS:3dimage/io

ref-mask =(input, io)
Reference image mask (binary). For supported file types see PLUGINS:3dimage/io

ref-mask-filter = ; factory
Filter to prepare the reference mask image, the output must be a binary image.. For supported plug-ins see PLUGINS:3dimage/filter

src =(input, io)
Study image. For supported file types see PLUGINS:3dimage/io

src-mask =(input, io)
Study image mask (binary). For supported file types see PLUGINS:3dimage/io

src-mask-filter = ; factory
Filter to prepare the study mask image, the output must be a binary image.. For supported plug-ins see PLUGINS:3dimage/filter

weight = 1; float
weight of cost function.

Evaluates the Sum of Squared Differences similarity measure by using three tagged image pairs. The cost function value is evaluated based on all image pairs, but the gradient is composed by composing its component based on the tag direction., supported parameters are:

refx =(input, io)
Reference image X-tag. For supported file types see PLUGINS:3dimage/io

refy =(input, io)
Reference image Y-tag. For supported file types see PLUGINS:3dimage/io

refz =(input, io)
Reference image Z-tag. For supported file types see PLUGINS:3dimage/io

srcx =(input, io)
Study image X-tag. For supported file types see PLUGINS:3dimage/io

srcy =(input, io)
Study image Y-tag. For supported file types see PLUGINS:3dimage/io

srcz =(input, io)
Study image Z-tag. For supported file types see PLUGINS:3dimage/io

weight = 1; float
weight of cost function.

Analyze 7.5 image

Recognized file extensions: .HDR, .hdr

Supported element types:
unsigned 8 bit, signed 16 bit, signed 32 bit, floating point 32 bit, floating point 64 bit

Virtual IO to and from the internal data pool

Recognized file extensions: .@

Dicom image series as 3D

Recognized file extensions: .DCM, .dcm

Supported element types:
signed 16 bit, unsigned 16 bit

HDF5 3D image IO

Recognized file extensions: .H5, .h5

Supported element types:
binary data, signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, signed 64 bit, unsigned 64 bit, floating point 32 bit, floating point 64 bit

INRIA image

Recognized file extensions: .INR, .inr

Supported element types:
signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, floating point 32 bit, floating point 64 bit

MetaIO 3D image IO using the VTK implementation (experimental).

Recognized file extensions: .MHA, .MHD, .mha, .mhd

Supported element types:
signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, floating point 32 bit, floating point 64 bit

NIFTI-1 3D image IO. The orientation is transformed in the same way like it is done with 'dicomtonifti --no-reorder' from the vtk-dicom package.

Recognized file extensions: .NII, .nii

Supported element types:
signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, signed 64 bit, unsigned 64 bit, floating point 32 bit, floating point 64 bit

VFF Sun raster format

Recognized file extensions: .VFF, .vff

Supported element types:
unsigned 8 bit, signed 16 bit

Vista 3D

Recognized file extensions: .-, .V, .VISTA, .v, .vista

Supported element types:
binary data, signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, floating point 32 bit, floating point 64 bit

3D image VTK-XML in- and output (experimental).

Recognized file extensions: .VTI, .vti

Supported element types:
signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, floating point 32 bit, floating point 64 bit

3D VTK image legacy in- and output (experimental).

Recognized file extensions: .VTK, .VTKIMAGE, .vtk, .vtkimage

Supported element types:
signed 8 bit, unsigned 8 bit, signed 16 bit, unsigned 16 bit, signed 32 bit, unsigned 32 bit, floating point 32 bit, floating point 64 bit

local normalized cross correlation with masking support., supported parameters are:

w = 5; uint in [1, 256]
half width of the window used for evaluating the localized cross correlation.

Spline parzen based mutual information with masking., supported parameters are:

cut = 0; float in [0, 40]
Percentage of pixels to cut at high and low intensities to remove outliers.

mbins = 64; uint in [1, 256]
Number of histogram bins used for the moving image.

mkernel = [bspline:d=3]; factory
Spline kernel for moving image parzen hinstogram. For supported plug-ins see PLUGINS:1d/splinekernel

rbins = 64; uint in [1, 256]
Number of histogram bins used for the reference image.

rkernel = [bspline:d=0]; factory
Spline kernel for reference image parzen hinstogram. For supported plug-ins see PLUGINS:1d/splinekernel

normalized cross correlation with masking support.

(no parameters)
Sum of squared differences with masking.

(no parameters)

18n
18n neighborhood 3D shape creator

(no parameters)
26n
26n neighborhood 3D shape creator

(no parameters)
6n
6n neighborhood 3D shape creator

(no parameters)
Closed spherical shape neighborhood including the pixels within a given radius r., supported parameters are:

r = 2; float in (0, inf)
sphere radius.

Affine transformation (12 degrees of freedom), supported parameters are:

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

Restricted rotation transformation (1 degrees of freedom). The transformation is restricted to the rotation around the given axis about the given rotation center, supported parameters are:

axis =(required, 3dfvector)
rotation axis.

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

origin =(required, 3dfvector)
center of the transformation.

Restricted affine transformation (3 degrees of freedom). The transformation is restricted to the rotation around the given axis and shearing along the two axis perpendicular to the given one, supported parameters are:

axis =(required, 3dfvector)
rotation axis.

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

origin =(required, 3dfvector)
center of the transformation.

Rigid transformation, i.e. rotation and translation (six degrees of freedom)., supported parameters are:

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

origin = [[0,0,0]]; 3dfvector
Relative rotation center, i.e. <0.5,0.5,0.5> corresponds to the center of the volume.

Rotation transformation (three degrees of freedom)., supported parameters are:

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

origin = [[0,0,0]]; 3dfvector
Relative rotation center, i.e. <0.5,0.5,0.5> corresponds to the center of the volume.

Restricted transformation (4 degrees of freedom). The transformation is restricted to the rotation around the x and y axis and a bending along the x axis, independedn in each direction, with the bending increasing with the squared distance from the rotation axis., supported parameters are:

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

norot = 0; bool
Don't optimize the rotation.

origin =(required, 3dfvector)
center of the transformation.

Free-form transformation that can be described by a set of B-spline coefficients and an underlying B-spline kernel., supported parameters are:

anisorate = [[0,0,0]]; 3dfvector
anisotropic coefficient rate in pixels, nonpositive values will be overwritten by the 'rate' value..

debug = 0; bool
enable additional debugging output.

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

kernel = [bspline:d=3]; factory
transformation spline kernel. For supported plug-ins see PLUGINS:1d/splinekernel

penalty = ; factory
transformation penalty energy term. For supported plug-ins see PLUGINS:3dtransform/splinepenalty

rate = 10; float in [1, inf)
isotropic coefficient rate in pixels.

Translation (three degrees of freedom), supported parameters are:

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

This plug-in implements a transformation that defines a translation for each point of the grid defining the domain of the transformation., supported parameters are:

imgboundary = mirror; factory
image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebc

imgkernel = [bspline:d=3]; factory
image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

Binary (non-portable) serialized IO of 3D transformations

Recognized file extensions: .bbs

Virtual IO to and from the internal data pool

Recognized file extensions: .@

Vista storage of 3D transformations

Recognized file extensions: .v, .v3dt

XML serialized IO of 3D transformations

Recognized file extensions: .x3dt

divcurl penalty on the transformation, supported parameters are:

curl = 1; float in [0, inf)
penalty weight on curl.

div = 1; float in [0, inf)
penalty weight on divergence.

norm = 0; bool
Set to 1 if the penalty should be normalized with respect to the image size.

weight = 1; float in (0, inf)
weight of penalty energy.

Gradient descent with automatic step size correction., supported parameters are:

ftolr = 0; double in [0, inf)
Stop if the relative change of the criterion is below..

max-step = 2; double in (0, inf)
Maximal absolute step size.

maxiter = 200; uint in [1, inf)
Stopping criterion: the maximum number of iterations.

min-step = 0.1; double in (0, inf)
Minimal absolute step size.

xtola = 0.01; double in [0, inf)
Stop if the inf-norm of the change applied to x is below this value..

Gradient descent with quadratic step estimation, supported parameters are:

ftolr = 0; double in [0, inf)
Stop if the relative change of the criterion is below..

gtola = 0; double in [0, inf)
Stop if the inf-norm of the gradient is below this value..

maxiter = 100; uint in [1, inf)
Stopping criterion: the maximum number of iterations.

scale = 2; double in (1, inf)
Fallback fixed step size scaling.

step = 0.1; double in (0, inf)
Initial step size.

xtola = 0; double in [0, inf)
Stop if the inf-norm of x-update is below this value..

optimizer plugin based on the multimin optimizers of the GNU Scientific Library (GSL) https://www.gnu.org/software/gsl/, supported parameters are:

eps = 0.01; double in (0, inf)
gradient based optimizers: stop when |grad| < eps, simplex: stop when simplex size < eps..

iter = 100; uint in [1, inf)
maximum number of iterations.

opt = gd; dict
Specific optimizer to be used.. Supported values are:
simplex ‐ Simplex algorithm of Nelder and Mead
cg-fr ‐ Flecher-Reeves conjugate gradient algorithm
cg-pr ‐ Polak-Ribiere conjugate gradient algorithm
bfgs ‐ Broyden-Fletcher-Goldfarb-Shann
bfgs2 ‐ Broyden-Fletcher-Goldfarb-Shann (most efficient version)
gd ‐ Gradient descent.

step = 0.001; double in (0, inf)
initial step size.

tol = 0.1; double in (0, inf)
some tolerance parameter.

Minimizer algorithms using the NLOPT library, for a description of the optimizers please see 'http://ab-initio.mit.edu/wiki/index.php/NLopt_Algorithms', supported parameters are:

ftola = 0; double in [0, inf)
Stopping criterion: the absolute change of the objective value is below this value.

ftolr = 0; double in [0, inf)
Stopping criterion: the relative change of the objective value is below this value.

higher = inf; double
Higher boundary (equal for all parameters).

local-opt = none; dict
local minimization algorithm that may be required for the main minimization algorithm.. Supported values are:
gn-direct ‐ Dividing Rectangles
gn-direct-l ‐ Dividing Rectangles (locally biased)
gn-direct-l-rand ‐ Dividing Rectangles (locally biased, randomized)
gn-direct-noscal ‐ Dividing Rectangles (unscaled)
gn-direct-l-noscal ‐ Dividing Rectangles (unscaled, locally biased)
gn-direct-l-rand-noscale ‐ Dividing Rectangles (unscaled, locally biased, randomized)
gn-orig-direct ‐ Dividing Rectangles (original implementation)
gn-orig-direct-l ‐ Dividing Rectangles (original implementation, locally biased)
ld-lbfgs-nocedal ‐ None
ld-lbfgs ‐ Low-storage BFGS
ln-praxis ‐ Gradient-free Local Optimization via the Principal-Axis Method
ld-var1 ‐ Shifted Limited-Memory Variable-Metric, Rank 1
ld-var2 ‐ Shifted Limited-Memory Variable-Metric, Rank 2
ld-tnewton ‐ Truncated Newton
ld-tnewton-restart ‐ Truncated Newton with steepest-descent restarting
ld-tnewton-precond ‐ Preconditioned Truncated Newton
ld-tnewton-precond-restart ‐ Preconditioned Truncated Newton with steepest-descent restarting
gn-crs2-lm ‐ Controlled Random Search with Local Mutation
ld-mma ‐ Method of Moving Asymptotes
ln-cobyla ‐ Constrained Optimization BY Linear Approximation
ln-newuoa ‐ Derivative-free Unconstrained Optimization by Iteratively Constructed Quadratic Approximation
ln-newuoa-bound ‐ Derivative-free Bound-constrained Optimization by Iteratively Constructed Quadratic Approximation
ln-neldermead ‐ Nelder-Mead simplex algorithm
ln-sbplx ‐ Subplex variant of Nelder-Mead
ln-bobyqa ‐ Derivative-free Bound-constrained Optimization
gn-isres ‐ Improved Stochastic Ranking Evolution Strategy
none ‐ don't specify algorithm

lower = -inf; double
Lower boundary (equal for all parameters).

maxiter = 100; int in [1, inf)
Stopping criterion: the maximum number of iterations.

opt = ld-lbfgs; dict
main minimization algorithm. Supported values are:
gn-direct ‐ Dividing Rectangles
gn-direct-l ‐ Dividing Rectangles (locally biased)
gn-direct-l-rand ‐ Dividing Rectangles (locally biased, randomized)
gn-direct-noscal ‐ Dividing Rectangles (unscaled)
gn-direct-l-noscal ‐ Dividing Rectangles (unscaled, locally biased)
gn-direct-l-rand-noscale ‐ Dividing Rectangles (unscaled, locally biased, randomized)
gn-orig-direct ‐ Dividing Rectangles (original implementation)
gn-orig-direct-l ‐ Dividing Rectangles (original implementation, locally biased)
ld-lbfgs-nocedal ‐ None
ld-lbfgs ‐ Low-storage BFGS
ln-praxis ‐ Gradient-free Local Optimization via the Principal-Axis Method
ld-var1 ‐ Shifted Limited-Memory Variable-Metric, Rank 1
ld-var2 ‐ Shifted Limited-Memory Variable-Metric, Rank 2
ld-tnewton ‐ Truncated Newton
ld-tnewton-restart ‐ Truncated Newton with steepest-descent restarting
ld-tnewton-precond ‐ Preconditioned Truncated Newton
ld-tnewton-precond-restart ‐ Preconditioned Truncated Newton with steepest-descent restarting
gn-crs2-lm ‐ Controlled Random Search with Local Mutation
ld-mma ‐ Method of Moving Asymptotes
ln-cobyla ‐ Constrained Optimization BY Linear Approximation
ln-newuoa ‐ Derivative-free Unconstrained Optimization by Iteratively Constructed Quadratic Approximation
ln-newuoa-bound ‐ Derivative-free Bound-constrained Optimization by Iteratively Constructed Quadratic Approximation
ln-neldermead ‐ Nelder-Mead simplex algorithm
ln-sbplx ‐ Subplex variant of Nelder-Mead
ln-bobyqa ‐ Derivative-free Bound-constrained Optimization
gn-isres ‐ Improved Stochastic Ranking Evolution Strategy
auglag ‐ Augmented Lagrangian algorithm
auglag-eq ‐ Augmented Lagrangian algorithm with equality constraints only
g-mlsl ‐ Multi-Level Single-Linkage (require local optimization and bounds)
g-mlsl-lds ‐ Multi-Level Single-Linkage (low-discrepancy-sequence, require local gradient based optimization and bounds)
ld-slsqp ‐ Sequential Least-Squares Quadratic Programming

step = 0; double in [0, inf)
Initial step size for gradient free methods.

stop = -inf; double
Stopping criterion: function value falls below this value.

xtola = 0; double in [0, inf)
Stopping criterion: the absolute change of all x-values is below this value.

xtolr = 0; double in [0, inf)
Stopping criterion: the relative change of all x-values is below this value.

Register image test.v to image ref.v by using a spline transformation with a coefficient rate of 5 and write the registered image to reg.v. Use two multiresolution levels, ssd as image cost function and divcurl weighted by 10.0 as transformation smoothness penalty.

mia-3dnonrigidreg -i test.v -r ref.v -o reg.v -l 2 -f spline:rate=3 image:cost=ssd divcurl:weight=10

Gert Wollny

This software is Copyright (c) 1999‐2015 Leipzig, Germany and Madrid, Spain. It comes with ABSOLUTELY NO WARRANTY and you may redistribute it under the terms of the GNU GENERAL PUBLIC LICENSE Version 3 (or later). For more information run the program with the option '--copyright'.

v2.4.7 USER COMMANDS