mia-3dnonrigidreg-alt(1)

📦
version 2.4.7 · 2 versions
, in debian bookworm, ubuntu jammy, ubuntu noble

mia-3dnonrigidreg-alt 💅

okay bestie, welcome to the manual page for mia-3dnonrigidreg-alt, living in the General Commands Manual section (section 1, obviously, iykyk) ✨

NAME

mia-3dnonrigidreg-alt — Non-linear registration of 3D images. 🧠✨

SYNOPSIS

mia-3dnonrigidreg-alt -o <out-transform> [options]
<PLUGINS:3dimage/fullcost>

DESCRIPTION

so mia-3dnonrigidreg-alt runs a non-rigid registration based on whatever cost criteria you hand it plus whatever transformation model you picked. 🦋

now here's the tea: unlike its sibling mia-3dnonrigidreg, this one does not support special command line parameters for feeding it the images. nope! instead, the images get specified directly when you define the cost function. which honestly? iconic. because of that, you can run image registrations that optimize the alignment of more than one image pair at the same time 🔥

⚠️ but lowkey important: all input images must be of the same dimension (in pixels). no mixing, no matching, periodt.

OPTIONS

  • -o, --out-transform — the output transformation. it's (output, required) and the type is io. for supported file types see PLUGINS:3dtransform/io 💾
  • -l, --levels — multi-resolution levels. default is 3. 🪜
  • -O, --optimizer — the optimizer used for minimization. default is gsl:opt=gd,step=0.1. for supported plugins see PLUGINS:minimizer/singlecost ⚙️
  • -f, --transForm — the transformation type (yes, that capital F is real, don't come for me). default is spline:rate=10. for supported plugins see PLUGINS:3dimage/transform 🌀

Help & Info 🫶

  • -V, --verbose — verbosity of output, prints messages of the given level and higher priorities. default is warning. the supported priorities, starting at the lowest level, are:
    1. trace — Function call trace
    2. debug — Debug output
    3. info — Low level messages
    4. message — Normal messages
    5. warning — Warnings
    6. fail — Report test failures
    7. error — Report errors
    8. fatal — Report only fatal errors
  • --copyright — prints the copyright information 📜
  • -h, --help — prints this help
  • -?, --usage — prints a short help
  • --version — prints the version number and exits 🏃‍♀️

Processing 🔥

  • --threads — maxiumum number of threads to use for processing (yes bestie, "maxiumum" is how it's written, we love her anyway 💀). this number should be lower or equal to the number of logical processor cores in the machine. default is -1, which means automatic estimation. 🧵

PLUGINS: 1d/spacialkernel ✨

  • cdiff — Central difference filter kernel, mirror boundary conditions are used. 🪞

(no parameters)

  • gauss — spacial Gauss filter kernel, supported parameters are:
    • w = 1; uint in [0, inf) — half filter width.
  • scharr — this plugin provides the 1D folding kernel for the Scharr gradient filter. 👑

(no parameters)

PLUGINS: 1d/splinebc 🌸

  • mirror — Spline interpolation boundary conditions that mirror on the boundary. 🪞

(no parameters)

  • repeat — Spline interpolation boundary conditions that repeats the value at the boundary. 🔁

(no parameters)

  • zero — Spline interpolation boundary conditions that assumes zero for values outside. 0️⃣

(no parameters)

PLUGINS: 1d/splinekernel 🎀

  • bspline — B-spline kernel creation, supported parameters are:
    • d = 3; int in [0, 5] — Spline degree.
  • omoms — OMoms-spline kernel creation, supported parameters are:
    • d = 3; int in [3, 3] — Spline degree.

PLUGINS: 3dimage/combiner 🦋

  • absdiff — Image combiner 'absdiff'

(no parameters)

  • add — Image combiner 'add'

(no parameters)

  • div — Image combiner 'div'

(no parameters)

  • mul — Image combiner 'mul'

(no parameters)

  • sub — Image combiner 'sub'

(no parameters)

PLUGINS: 3dimage/cost 💖

  • lncc — 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.
  • mi — 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 histogram. 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 histogram. for supported plug-ins see PLUGINS:1d/splinekernel
  • ncc — normalized cross correlation. 📐

(no parameters)

  • ngf — 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). the supported values are:
      • ds — square of scaled difference
      • dot — scalar product kernel
      • cross — cross product kernel
  • ssd — 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.
  • ssd-automask — 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.

PLUGINS: 3dimage/filter 🌈

  • bandpass — intensity bandpass filter, supported parameters are:
    • max = 3.40282e+38; float — maximum of the band.
    • min = 0; float — minimum of the band.
  • binarize — image binarize filter, supported parameters are:
    • max = 3.40282e+38; float — maximum of accepted range.
    • min = 0; float — minimum of accepted range.
  • close — 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
  • combiner — 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 =(input, required, 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.
  • convert — 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. the 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. the 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 — 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.
  • dilate — 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
  • distance — 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 🫧

Bestie, this one shrinks your image: you hand it a block size, and that block size is your downscale factor. But first — and this is the glow-up — it filters the image with a smoothing filter to kill off the high frequency data so you don't get aliasing artifacts. ✨

the parameters are giving:

  • b = [[1,1,1]]; 3dbounds — the 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 — the smoothing filter kernel that gets applied, and the size of that filter is estimated from the blocksize. For supported plug-ins see PLUGINS:1d/spacialkernel

erode 🪨

a 3d image stack erode filter, no cap. params:

  • hint = black; string — a hint about what the main image content is (black|white).
  • shape = [sphere:r=2]; factory — the structuring element. For supported plug-ins see PLUGINS:3dimage/shape

gauss 🌫️

isotropic 3D gauss filter, slay. params:

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

gradnorm 📐

3D image to gradient norm filter.

  • (no parameters) — she's low maintenance 💅

growmask 🌱

take an input binary mask plus a reference gray scale image and do region growing: you add the neighborhood pixels of a pixel that's already been added, as long as they have a lower intensity that's still above the given threshold. it's giving fungus growth but make it medical imaging. 🍄

params:

  • min = 1; float — lower threshold for mask growing.
  • ref =(input, required, 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

invert 🔄

intensity invert filter. that's it, that's the whole vibe.

  • (no parameters)

isovoxel 🧊

this filter scales an image so that the voxel size becomes isometric and its size matches the value you give it. iconic behavior honestly.

params:

  • 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.

kmeans 🎯

3D image k-means filter. in the output image the pixel value represents the class membership, and the class centers get stored as an attribute in the image. cute AND informative.

params:

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

label 🏷️

a filter to label the connected components of a binary image.

params:

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

labelmap 🗺️

image filter to remap label id's. only applicable to images with integer valued intensities/labels — so don't try this on your float data bestie, she will not cooperate.

params:

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

labelscale 🗳️

a filter that only creates output voxels that already exist in the input image. scaling happens via a voting algorithm that picks the target pixel value based on the highest pixel count of a certain label in the corresponding source region. and if that region has two labels tied with the same count? the one with the lower number wins. democracy but make it labels. 🫡

params:

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

load 📂

load the input image from a file and use it to replace the current image in the pipeline. she just swaps in mid-flow.

params:

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

lvdownscale 🗳️✨

this is a label voting downscale filter. it downscales a 3D image by blocks. for each block, the (non-zero) label that appears most times in that block gets issued as the output pixel in the target image. and if two labels show up the same number of times? the one with the lower absolute value wins. periodt.

params:

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

mask 🎭

mask an image: one image comes from the parameters list and the other comes from the normal filter input. both images must be the same dimensions and one of them must be binary — that's not a suggestion, that's the rules. 🚨 the attributes of the image coming through the filter pipeline are preserved, and the output pixel type corresponds to whichever input image is not binary.

params:

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

mean ➗

3D image mean filter.

params:

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

median 🎚️

median 3d filter.

params:

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

mlv 🧠

Mean of Least Variance 3D image filter.

params:

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

msnormalizer 📊

3D image mean-sigma normalizing filter.

params:

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

open 🌸

morphological open.

params:

  • 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

reorient 🧭

3D image reorientation filter. spin that volume around, bestie.

params:

  • 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 📏

resize an image. the original data is centered within the new sized image, so nothing gets shoved into a corner. 🫶

params:

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

sandp 🧂🌶️

salt and pepper 3d filter. season your image, literally.

params:

  • thresh = 100; float in [0, inf) — thresh value.
  • w = 1; int in [1, inf) — filter width parameter.

scale 🎈

3D image filter that scales to a given target size .

params:

  • 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).

scharr 🌈

the 3D Scharr filter for gradient evaluation. ⚠️ heads up: 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. don't say I didn't warn you.

params:

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

selectbig 👑

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 — so yeah, chaotic, proceed with caution 💀. The input pixel must have an integral pixel type.

  • (no parameters)

sepconv 🪄

3D image intensity separaple convolution filter. three kernels, three directions, one gorgeous separable moment.

params:

  • 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

sobel 🔲

the 2D Sobel filter for gradient evaluation. ⚠️ same disclaimer as her sister above: 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.

params:

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

sws

okay so this one is the seeded watershed — seeded meaning you tell it where to start. 💧 The algorithm extracts exactly as many regions as initial labels are given in the seed image. no cap, your seed image is literally the blueprint.

Parameters, bestie:

  • grad = 0; bool — set this and the input image gets read as a gradient. that's it, that's the job.
  • mark = 0; bool — turns on a special gray scale value so the segmented watersheds get marked. ✨
  • n = [sphere:r=1]; factory — the neighborhood for watershed region growing. want the plug-ins that work here? go look at PLUGINS:3dimage/shape. 👀
  • seed =(input, required, string) — the seed input image, aka the one holding the labels for the initial regions. required, so no skipping it, babe. 🎀

tee

saves the input image to a file and passes it straight through to the next filter. it's giving "have your cake and eat it too" energy. 🍓

  • file =(output, required, io) — name of the output file to save the image to. supported file types live over at PLUGINS:3dimage/io.

thinning

3D morphological thinning, and it's built on some real ones: Lee and Kashyap, 'Building Skeleton Models via 3-D Medial Surface/Axis Thinning Algorithms', Graphical Models and Image Processing, 56(6):462-478, 1994. 📚

heads up: this implementation only supports the 26 neighbourhood. don't come at it with anything else.
  • (no parameters) — yeah, zero knobs. she just works. 💅

transform

transforms the input image with the given transformation. iconic. 🔄

  • file =(input, required, io) — name of the file containing the transformation. required! file types are listed in PLUGINS:3dtransform/io.
  • imgboundary = ; factory — override the image interpolation boundary conditions. plug-ins for this live in PLUGINS:1d/splinebc.
  • imgkernel = ; factory — override the image interpolator kernel. plug-ins for this one live in PLUGINS:1d/splinekernel.

variance

3D image variance filter. 🌈

  • w = 1; int in [1, inf) — half filter width. that's the whole parameter list, she's low maintenance.

ws

basic watershed segmentation. the O.G. no seeds, just vibes. 💦

  • evalgrad = 0; bool — set it to 1 if your input image does not represent a gradient norm image. otherwise leave her alone.
  • mark = 0; bool — same deal as before: marks the segmented watersheds with a special gray scale value. ✨
  • n = [sphere:r=1]; factory — neighborhood for watershed region growing. supported plug-ins? PLUGINS:3dimage/shape, bestie.
  • thresh = 0; float in [0, 1) — relative gradient norm threshold. the actual threshold value is thresh * (max_grad - min_grad) + min_grad. basins separated by gradients with a lower norm will be joined. 👑

PLUGINS: 3dimage/fullcost

image

a generalized image similarity cost function that also handles multi-resolution processing. the actual similarity measure is given as an extra parameter, so you're never locked in. 🦋

  • cost = ssd; factory — the cost function kernel. plug-ins for days over at PLUGINS:3dimage/cost.
  • debug = 0; bool — saves intermediate results so you can debug. lowkey a lifesaver when things go sideways. 😭
  • ref =(input, io) — the reference image. file types: PLUGINS:3dimage/io.
  • src =(input, io) — the study image. file types: PLUGINS:3dimage/io.
  • weight = 1; float — weight of the cost function. yes she's just a float, and yes she matters. 💖

labelimage

a similarity cost function that maps labels of two images and handles label-preserving multi-resolution processing. 🏷️

  • maxlabel = 256; int in [2, 32000] — the maximum number of labels to consider. she's capped at 32000, don't push her.
  • ref =(input, io) — reference image. file types: PLUGINS:3dimage/io.
  • src =(input, io) — study image. file types: PLUGINS:3dimage/io.
  • weight = 1; float — weight of the cost function.

maskedimage

a generalized masked image similarity cost function that also handles multi-resolution processing. 🎭

okay real talk, important caveat: the provided masks should be densely filled regions in multi-resolution processing, because otherwise the mask information may get lost when downscaling the image. the mask may be pre-filtered — but 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 as an extra parameter.

  • cost = ssd; factory — cost function kernel. plug-ins over at PLUGINS:3dimage/maskedcost.
  • ref =(input, io) — reference image. file types: PLUGINS:3dimage/io.
  • ref-mask =(input, io) — reference image mask (binary). file types: PLUGINS:3dimage/io.
  • ref-mask-filter = ; factory — filter to prepare the reference mask image, and the output must be a binary image. plug-ins: PLUGINS:3dimage/filter.
  • src =(input, io) — study image. file types: PLUGINS:3dimage/io.
  • src-mask =(input, io) — study image mask (binary). file types: PLUGINS:3dimage/io.
  • src-mask-filter = ; factory — filter to prepare the study mask image, and again the output must be a binary image. plug-ins: PLUGINS:3dimage/filter.
  • weight = 1; float — weight of the cost function. 💜

taggedssd

evaluates the Sum of Squared Differences similarity measure by using three tagged image pairs. 🏷️🏷️🏷️ here's the twist though: 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. so she's mixing and matching, and she knows exactly what she's doing.

  • refx =(input, io) — reference image X-tag. file types: PLUGINS:3dimage/io.
  • refy =(input, io) — reference image Y-tag. file types: PLUGINS:3dimage/io.
  • refz =(input, io) — reference image Z-tag. file types: PLUGINS:3dimage/io.
  • srcx =(input, io) — study image X-tag. file types: PLUGINS:3dimage/io.
  • srcy =(input, io) — study image Y-tag. file types: PLUGINS:3dimage/io.
  • srcz =(input, io) — study image Z-tag. file types: PLUGINS:3dimage/io.
  • weight = 1; float — weight of the cost function. ⭐

six images, three tags, one iconic cost function. we're not done yet though, there's more coming, stay tuned bestie 🫶

🦋 PLUGINS: 3dimage/io 🦋

okay bestie this is the whole reader/writer squad — every one of these cuties knows how to pull a 3D image in and push one back out, and they each claim their own file extensions like little territory flags 💅

analyze — Analyze 7.5 image 🧠

The classic. Old-school but iconic, fr fr.

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

datapool — Virtual IO to and from the internal data pool 🫧

Not a file, not a vibe, it's virtual IO straight to and from the internal data pool. lowkey iconic.

Recognized file extensions: .@

dicom — Dicom image series as 3D 🩻

Takes a whole DICOM series and says "this is one 3D image now, periodt."

Recognized file extensions: .DCM, .dcm

Supported element types:

  • signed 16 bit
  • unsigned 16 bit

hdf5 — 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 — 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

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

heads up bestie, this one is experimental — it's giving "we're still figuring it out" energy. no cap.

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 — 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. yes, exactly like that, don't come for me.

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 — VFF Sun raster format ☀️

Recognized file extensions: .VFF, .vff

Supported element types:

  • unsigned 8 bit
  • signed 16 bit

vista — 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

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

another experimental one, bestie. proceed with love and a backup. 🙈

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

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

legacy AND experimental? it's giving vintage-but-also-unstable, and honestly? respect. 💀

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

🎀 PLUGINS: 3dimage/maskedcost 🎀

this is the cost function crew, but make it masked. every one of them knows how to ignore the pixels you told it to ignore, slay. 🫶

lncc — 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. so like, the neighborhood it peeks at is built from this. 🌈

mi — 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. bye bye extremes, you will not be missed. ✂️
  • 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 👑

ncc — normalized cross correlation with masking support. 💖

(no parameters)

so unbothered. so effortless. no knobs to turn, just vibes.

ssd — Sum of squared differences with masking. 🧮

(no parameters)

also zero parameters, also zero drama.

⭐ PLUGINS: 3dimage/shape ⭐

shape makers!! these little guys build neighborhoods for you, and each one has its own whole personality.

18n — 18n neighborhood 3D shape creator 🎲

(no parameters)

26n — 26n neighborhood 3D shape creator 🎲

(no parameters)

6n — 6n neighborhood 3D shape creator 🎲

(no parameters)

sphere — Closed spherical shape neighborhood including the pixels within a given radius r., supported parameters are: 🔮

  • r = 2; float in (0, inf) — sphere radius. it goes on forever bestie, (0, inf) fr fr, just pick your size. 🌸

🦋 PLUGINS: 3dimage/transform 🦋

okay bestie this is THE section, the one where the actual moving-around-magic lives. every one of these is a transformation plug-in, and they all share some fam parameters (hi imgboundary and imgkernel, we know you), so watch for the repeats, they're not typos, they're just how it is 💅

affine</p> — Affine transformation (12 degrees of freedom) 👑

the full-send one, 12 degrees of freedom, it can do a LOT. supported parameters are:

  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.

axisrot — Restricted rotation transformation (1 degrees of freedom) 🌸

a lil baby rotation, only 1 degree of freedom. it's locked to rotating 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 the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.
  • origin =(required, 3dfvector) — center of the transformation.

raffine — Restricted affine transformation (3 degrees of freedom) ✨

affine but on a leash: 3 degrees of freedom. it's restricted to rotating around the given axis plus shearing along the two axis perpendicular to it. supported parameters are:

  • axis =(required, 3dfvector) — rotation axis.
  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.
  • origin =(required, 3dfvector) — center of the transformation.

rigid — Rigid transformation 🫶

rotation and translation together, six degrees of freedom. yes bestie, the original really does put a stray period before the comma there, we're leaving it alone, not our business. supported parameters are:

  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at 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 — Rotation transformation (three degrees of freedom) 🌀

just rotation, three degrees of freedom, nothing extra. supported parameters are:

  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at 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. 🎯

rotbend — Restricted transformation (4 degrees of freedom) 🌈

4 degrees of freedom, and it's a whole vibe: rotation around the x and y axis, plus a bending along the x axis, independedn in each direction (that's the original's spelling, she's quirky, we love her), with the bending increasing with the squared distance from the rotation axis. supported parameters are:

  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.
  • norot = 0 (bool) — don't optimize the rotation.
  • origin =(required, 3dfvector) — center of the transformation.

spline — Free-form transformation 🧚

the free-form icon, the main character fr. it 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. no cap.
  • debug = 0 (bool) — enable additional debugging output. 🙈
  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.
  • kernel = [bspline:d=3] (factory) — transformation spline kernel. supported plug-ins live at PLUGINS:1d/splinekernel.
  • penalty = (factory) — transformation penalty energy term. supported plug-ins live at PLUGINS:3dtransform/splinepenalty.
  • rate = 10 (float in [1, inf)) — isotropic coefficient rate in pixels.
👀 heads up: penalty has no default value, it's literally just empty before the ;. and anisorate's nonpositive values get clobbered by rate, so don't be out here setting negative numbers and expecting them to stick.

translate — Translation (three degrees of freedom) 🚚

just slides the whole thing, three degrees of freedom, honest work. supported parameters are:

  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.

vf — vector field transformation 🌊

this plug-in implements a transformation that defines a translation for each point of the grid defining the domain of the transformation. like, every point gets its own lil translation, bespoke. supported parameters are:

  • imgboundary = mirror (factory) — image interpolation boundary conditions. for the plug-ins that play nice here, go look at PLUGINS:1d/splinebc.
  • imgkernel = [bspline:d=3] (factory) — image interpolator kernel. supported plug-ins live at PLUGINS:1d/splinekernel.

💾 PLUGINS: 3dtransform/io 💾

the ones that read and write your 3D transformations. save your work bestie, always. 🫶

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

  • Recognized file extensions: .bbs
⚠️ it says non-portable right there in the name, so don't be shocked when it doesn't vibe with another machine.

datapool — Virtual IO to and from the internal data pool

  • Recognized file extensions: .@

vista — Vista storage of 3D transformations

  • Recognized file extensions: .v, .v3dt

xml — XML serialized IO of 3D transformations

  • Recognized file extensions: .x3dt

🎀 PLUGINS: 3dtransform/splinepenalty 🎀

the penalty section. this is where you tell the spline "no, bad, stop stretching like that" ✋

divcurl — 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.
note the brackets bestie: curl and div are [0, inf) — zero is allowed. weight is (0, inf) — zero is not invited to this party, strictly greater than zero only. 💀

🌸 gdas 🌸

gradient descent but it auto-corrects its own step size, so it's giving self-aware queen energy 💅

  • ftolr = 0; double in [0, inf) — bail out when the relative change of the criterion drops below this.
  • max-step = 2; double in (0, inf) — the biggest absolute step size it's allowed to take.
  • maxiter = 200; uint in [1, inf) — stopping criterion: the maximum number of iterations, periodt.
  • min-step = 0.1; double in (0, inf) — the tiniest absolute step size.
  • xtola = 0.01; double in [0, inf) — bail out when the inf-norm of the change applied to x drops below this.

🦋 gdsq 🦋

gradient descent with quadratic step estimation, lowkey smarter than plain gd ✨

  • ftolr = 0; double in [0, inf) — bail out when the relative change of the criterion drops below this.
  • gtola = 0; double in [0, inf) — bail out when the inf-norm of the gradient drops below this.
  • 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) — bail out when the inf-norm of the x-update drops below this.

🧚 gsl 🧚

an optimizer plugin riding on the multimin optimizers from the GNU Scientific Library (GSL) https://www.gnu.org/software/gsl/. iconic library, no cap.

  • eps = 0.01; double in (0, inf) — for gradient based optimizers: stop when |grad| < eps; for simplex: stop when the simplex size < eps.
  • iter = 100; uint in [1, inf) — maximum number of iterations.
  • opt = gd; dict — which specific optimizer to use. 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.

🌈 nlopt 🌈

minimizer algorithms via the NLOPT library. if you wanna read the whole tea on these optimizers, it's at 'http://ab-initio.mit.edu/wiki/index.php/NLopt_Algorithms'. the 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 — the 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 — 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
    • 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: the 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.

🎀 EXAMPLE 🎀

so bestie, say you wanna register image test.v onto image ref.v using a spline transformation with a coefficient rate of 5, and write the registered image out to reg.v. you want two multiresolution levels, ssd as the image cost function, and divcurl weighted by 10.0 as the transformation smoothness penalty. the resulting transformation gets saved in reg.vf. here's the whole vibe:

mia-3dnonrigidreg-alt -o reg.vf -l 2 -f spline:rate=3
       image:cost=ssd,src=test.v,ref=ref.v divcurl:weight=10

👑 AUTHOR(s) 👑

Gert Wollny — the actual main character who wrote this 🫶

okay bestie, the legal bit, but make it cute ✨ this software is Copyright (c) 1999‐2015 Leipzig, Germany and Madrid, Spain. two cities, one iconic codebase, we love to see it 🫶

and now the part where the lawyers get loud: it comes with ABSOLUTELY NO WARRANTY. like, none. zero. nada. it's giving "you're on your own, bestie" 😭 so pls don't come for anybody if it does a little something silly.

but the good news?? you may redistribute it under the terms of the GNU GENERAL PUBLIC LICENSE Version 3 (or later). so go forth, share it, remix it, be free 💅 just respect the license, periodt.

  • want the full deets? for more information run the program with the option --copyright 📜

and that's the whole page, bestie!! 🎀 this was mia-3dnonrigidreg-alt(1) from the USER COMMANDS manual, v2.4.7 — check that lil date stamp down at the bottom too, that's when this beauty got written up 👑 you made it all the way through all seven parts, which is main character energy fr fr 🌈💖 go register something nonrigidly, you've earned it. no cap. byeee 🦋✨

📦 where this page comes from