Provided by: mia-tools_2.4.6-4ubuntu2_amd64

**NAME**

('mia\-3dnonrigidreg\-alt',) - Non-linear registration of 3D images.

**SYNOPSIS**

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

**DESCRIPTION**

mia-3dnonrigidreg-altThis program runs a non-rigid registration based on the given cost criteria and a given transformation model. Other than mia-3dnonrigidreg it doesn't support specific command line parameters to provide the images. Instead the images are specified dirctly when defining the cost function. Hence, image registrations can be executed that optimize the aligmnet of more than one image pair at the same time. Note, however, that all input images must be of the same dimension (in pixels)

**OPTIONS**

-o --out-transform=(output, required); io output transformation For supported file types see PLUGINS:3dtransform/io -l --levels=3 multi-resolution levels -O --optimizer=gsl:opt=gd,step=0.1 Optimizer used for minimization For supported plugins see PLUGINS:minimizer/singlecost -f --transForm=spline:rate=10 transformation type For supported plugins see PLUGINS:3dimage/transformHelp&Info-V --verbose=warning verbosity of output, print messages of given level and higher priorities. Supported priorities starting at lowest level are:info‐ Low level messagestrace‐ Function call tracefail‐ Report test failureswarning‐ Warningserror‐ Report errorsdebug‐ Debug outputmessage‐ Normal messagesfatal‐ Report only fatal errors --copyright print copyright information -h --help print this help -? --usage print a short help --version print the version number and exitProcessing--threads=-1 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).

**PLUGINS:** **1d/spacialkernel**

cdiffCentral difference filter kernel, mirror boundary conditions are used. (no parameters)gaussspacial Gauss filter kernel, supported parameters are:w= 1; uint in [0, inf) half filter width.scharrThis plugin provides the 1D folding kernel for the Scharr gradient filter (no parameters)

**PLUGINS:** **1d/splinebc**

mirrorSpline interpolation boundary conditions that mirror on the boundary (no parameters)repeatSpline interpolation boundary conditions that repeats the value at the boundary (no parameters)zeroSpline interpolation boundary conditions that assumes zero for values outside (no parameters)

**PLUGINS:** **1d/splinekernel**

bsplineB-spline kernel creation , supported parameters are:d= 3; int in [0, 5] Spline degree.omomsOMoms-spline kernel creation, supported parameters are:d= 3; int in [3, 3] Spline degree.

**PLUGINS:** **3dimage/combiner**

absdiffImage combiner 'absdiff' (no parameters)addImage combiner 'add' (no parameters)divImage combiner 'div' (no parameters)mulImage combiner 'mul' (no parameters)subImage combiner 'sub' (no parameters)

**PLUGINS:** **3dimage/cost**

lncclocal 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.miSpline 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/splinekernelrbins= 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/splinekernelnccnormalized cross correlation. (no parameters)ngfThis 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 differencedot‐ scalar product kernelcross‐ cross product kernelssd3D 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-automask3D 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**

bandpassintensity bandpass filter, supported parameters are:max= 3.40282e+38; float maximum of the band.min= 0; float minimum of the band.binarizeimage binarize filter, supported parameters are:max= 3.40282e+38; float maximum of accepted range.min= 0; float minimum of accepted range.closemorphological 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/shapecombinerCombine 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/ioop=(required, factory) Image combiner to be applied to the images. For supported plug-ins see PLUGINS:3dimage/combinerreverse= 0; bool reverse the order in which the images passed to the combiner.convertimage 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:opt‐ apply a linear transformation that maps the real input range to the full output rangerange‐ apply linear transformation that maps the input data type range to the output data type rangecopy‐ copy data when convertinglinear‐ apply linear transformation x -> a*x+boptstat‐ apply a linear transform that maps based on input mean and variation to the full output rangerepn= ubyte; dict output pixel type. Supported values are:none‐ no pixel type definedfloat‐ floating point 32 bitsbyte‐ signed 8 bitulong‐ unsigned 64 bitdouble‐ floating point 64 bitsint‐ signed 32 bitushort‐ unsigned 16 bitsshort‐ signed 16 bituint‐ unsigned 32 bitslong‐ signed 64 bitbit‐ binary dataubyte‐ unsigned 8 bitcropCrop 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.dilate3d 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/shapedistanceEvaluate 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)downscaleDownscale 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/spacialkernelerode3d 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/shapegaussisotropic 3D gauss filter, supported parameters are:w= 1; int in [0, inf) filter width parameter.gradnorm3D image to gradient norm filter (no parameters)growmaskUse 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=(input, required, io) reference image for mask region growing. For supported file types see PLUGINS:3dimage/ioshape= 6n; factory neighborhood mask. For supported plug-ins see PLUGINS:3dimage/shapeinvertintensity invert filter (no parameters)isovoxelThis 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/splinekernelsize= 1; float in (0, inf) isometric target voxel size.kmeans3D 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.labelA 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/shapelabelmapImage filter to remap label id's. Only applicable to images with integer valued intensities/labels., supported parameters are:map=(input, required, string) Label mapping file.labelscaleA 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.loadLoad the input image from a file and use it to replace the current image in the pipeline., supported parameters are:file=(input, required, io) name of the input file to load from.. For supported file types see PLUGINS:3dimage/iolvdownscaleThis 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..maskMask 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=(input, required, io) second input image file name. For supported file types see PLUGINS:3dimage/iomean3D image mean filter, supported parameters are:w= 1; int in [1, inf) half filter width.medianmedian 3d filter, supported parameters are:w= 1; int in [1, inf) filter width parameter.mlvMean of Least Variance 3D image filter, supported parameters are:w= 1; int in [1, inf) filter width parameter.msnormalizer3D image mean-sigma normalizing filter, supported parameters are:w= 1; int in [1, inf) half filter width.openmorphological 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/shapereorient3D image reorientation filter, supported parameters are:map= xyz; dict oriantation mapping to be applied. Supported values are:p-zxy‐ permutate x->y->z->xr-x180‐ rotate around x-axis clockwise 180 degreexyz‐ keep orientationp-yzx‐ permutate x->z->y->xr-z180‐ rotate around z-axis clockwise 180 degreer-y270‐ rotate around y-axis clockwise 270 degreef-xz‐ flip x-zf-yz‐ flip y-zr-x90‐ rotate around x-axis clockwise 90 degreer-y90‐ rotate around y-axis clockwise 90 degreer-x270‐ rotate around x-axis clockwise 270 degreer-z270‐ rotate around z-axis clockwise 270 degreer-z90‐ rotate around z-axis clockwise 90 degreef-xy‐ flip x-yr-y180‐ rotate around y-axis clockwise 180 degreeresizeResize 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..sandpsalt and pepper 3d filter, supported parameters are:thresh= 100; float in [0, inf) thresh value.w= 1; int in [1, inf) filter width parameter.scale3D 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/splinekernels= [[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).scharrThe 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:y‐ gradient in y-directionx‐ gradient in x-directionz‐ gradient in z-directionselectbigA 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)sepconv3D 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/spacialkernelky= [gauss:w=1]; factory filter kernel in y-direction. For supported plug-ins see PLUGINS:1d/spacialkernelkz= [gauss:w=1]; factory filter kernel in z-direction. For supported plug-ins see PLUGINS:1d/spacialkernelsobelThe 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:y‐ gradient in y-directionx‐ gradient in x-directionz‐ gradient in z-directionswsseeded 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/shapeseed=(input, required, string) seed input image containing the lables for the initial regions.teeSave the input image to a file and also pass it through to the next filter, supported parameters are:file=(output, required, io) name of the output file to save the image too.. For supported file types see PLUGINS:3dimage/iothinning3D 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)transformTransform the input image with the given transformation., supported parameters are:file=(input, required, io) Name of the file containing the transformation.. For supported file types see PLUGINS:3dtransform/ioimgboundary= ; factory override image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebcimgkernel= ; factory override image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelvariance3D image variance filter, supported parameters are:w= 1; int in [1, inf) half filter width.wsbasic 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/shapethresh= 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.

**PLUGINS:** **3dimage/fullcost**

imageGeneralized 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/costdebug= 0; bool Save intermediate resuts for debugging.ref=(input, io) Reference image. For supported file types see PLUGINS:3dimage/iosrc=(input, io) Study image. For supported file types see PLUGINS:3dimage/ioweight= 1; float weight of cost function.labelimageSimilarity 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/iosrc=(input, io) Study image. For supported file types see PLUGINS:3dimage/ioweight= 1; float weight of cost function.maskedimageGeneralized 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/maskedcostref=(input, io) Reference image. For supported file types see PLUGINS:3dimage/ioref-mask=(input, io) Reference image mask (binary). For supported file types see PLUGINS:3dimage/ioref-mask-filter= ; factory Filter to prepare the reference mask image, the output must be a binary image.. For supported plug-ins see PLUGINS:3dimage/filtersrc=(input, io) Study image. For supported file types see PLUGINS:3dimage/iosrc-mask=(input, io) Study image mask (binary). For supported file types see PLUGINS:3dimage/iosrc-mask-filter= ; factory Filter to prepare the study mask image, the output must be a binary image.. For supported plug-ins see PLUGINS:3dimage/filterweight= 1; float weight of cost function.taggedssdEvaluates 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/iorefy=(input, io) Reference image Y-tag. For supported file types see PLUGINS:3dimage/iorefz=(input, io) Reference image Z-tag. For supported file types see PLUGINS:3dimage/iosrcx=(input, io) Study image X-tag. For supported file types see PLUGINS:3dimage/iosrcy=(input, io) Study image Y-tag. For supported file types see PLUGINS:3dimage/iosrcz=(input, io) Study image Z-tag. For supported file types see PLUGINS:3dimage/ioweight= 1; float weight of cost function.

**PLUGINS:** **3dimage/io**

analyzeAnalyze 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 bitdatapoolVirtual IO to and from the internal data pool ('Recognized file extensions: ', '.@')dicomDicom image series as 3D ('Recognized file extensions: ', '.DCM, .dcm') Supported element types: signed 16 bit, unsigned 16 bithdf5HDF5 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 bitinriaINRIA 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 bitmhdMetaIO 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 bitniftiNIFTI-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 bitvffVFF Sun raster format ('Recognized file extensions: ', '.VFF, .vff') Supported element types: unsigned 8 bit, signed 16 bitvistaVista 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 bitvti3D 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 bitvtk3D 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

**PLUGINS:** **3dimage/maskedcost**

lncclocal 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.miSpline 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/splinekernelrbins= 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/splinekernelnccnormalized cross correlation with masking support. (no parameters)ssdSum of squared differences with masking. (no parameters)

**PLUGINS:** **3dimage/shape**

18n18n neighborhood 3D shape creator (no parameters)26n26n neighborhood 3D shape creator (no parameters)6n6n neighborhood 3D shape creator (no parameters)sphereClosed spherical shape neighborhood including the pixels within a given radius r., supported parameters are:r= 2; float in (0, inf) sphere radius.

**PLUGINS:** **3dimage/transform**

affineAffine transformation (12 degrees of freedom), supported parameters are:imgboundary= mirror; factory image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelaxisrotRestricted 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/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelorigin=(required, 3dfvector) center of the transformation.raffineRestricted 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/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelorigin=(required, 3dfvector) center of the transformation.rigidRigid 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/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelorigin= [[0,0,0]]; 3dfvector Relative rotation center, i.e. <0.5,0.5,0.5> corresponds to the center of the volume.rotationRotation transformation (three degrees of freedom)., supported parameters are:imgboundary= mirror; factory image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelorigin= [[0,0,0]]; 3dfvector Relative rotation center, i.e. <0.5,0.5,0.5> corresponds to the center of the volume.rotbendRestricted 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/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelnorot= 0; bool Don't optimize the rotation.origin=(required, 3dfvector) center of the transformation.splineFree-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/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelkernel= [bspline:d=3]; factory transformation spline kernel. For supported plug-ins see PLUGINS:1d/splinekernelpenalty= ; factory transformation penalty energy term. For supported plug-ins see PLUGINS:3dtransform/splinepenaltyrate= 10; float in [1, inf) isotropic coefficient rate in pixels.translateTranslation (three degrees of freedom), supported parameters are:imgboundary= mirror; factory image interpolation boundary conditions. For supported plug-ins see PLUGINS:1d/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernelvfThis 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/splinebcimgkernel= [bspline:d=3]; factory image interpolator kernel. For supported plug-ins see PLUGINS:1d/splinekernel

**PLUGINS:** **3dtransform/io**

bbsBinary (non-portable) serialized IO of 3D transformations ('Recognized file extensions: ', '.bbs')datapoolVirtual IO to and from the internal data pool ('Recognized file extensions: ', '.@')vistaVista storage of 3D transformations ('Recognized file extensions: ', '.v, .v3dt')xmlXML serialized IO of 3D transformations ('Recognized file extensions: ', '.x3dt')

**PLUGINS:** **3dtransform/splinepenalty**

divcurldivcurl 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.

**PLUGINS:** **minimizer/singlecost**

gdasGradient 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..gdsqGradient 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..gsloptimizer 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:bfgs‐ Broyden-Fletcher-Goldfarb-Shannbfgs2‐ Broyden-Fletcher-Goldfarb-Shann (most efficient version)cg-fr‐ Flecher-Reeves conjugate gradient algorithmgd‐ Gradient descent.simplex‐ Simplex algorithm of Nelder and Meadcg-pr‐ Polak-Ribiere conjugate gradient algorithmstep= 0.001; double in (0, inf) initial step size.tol= 0.1; double in (0, inf) some tolerance parameter.nloptMinimizer 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-orig-direct-l‐ Dividing Rectangles (original implementation, locally biased)gn-direct-l-noscal‐ Dividing Rectangles (unscaled, locally biased)gn-isres‐ Improved Stochastic Ranking Evolution Strategyld-tnewton‐ Truncated Newtongn-direct-l-rand‐ Dividing Rectangles (locally biased, randomized)ln-newuoa‐ Derivative-free Unconstrained Optimization by Iteratively Constructed Quadratic Approximationgn-direct-l-rand-noscale‐ Dividing Rectangles (unscaled, locally biased, randomized)gn-orig-direct‐ Dividing Rectangles (original implementation)ld-tnewton-precond‐ Preconditioned Truncated Newtonld-tnewton-restart‐ Truncated Newton with steepest-descent restartinggn-direct‐ Dividing Rectanglesln-neldermead‐ Nelder-Mead simplex algorithmln-cobyla‐ Constrained Optimization BY Linear Approximationgn-crs2-lm‐ Controlled Random Search with Local Mutationld-var2‐ Shifted Limited-Memory Variable-Metric, Rank 2ld-var1‐ Shifted Limited-Memory Variable-Metric, Rank 1ld-mma‐ Method of Moving Asymptotesld-lbfgs-nocedal‐ Noneld-lbfgs‐ Low-storage BFGSgn-direct-l‐ Dividing Rectangles (locally biased)none‐ don't specify algorithmln-bobyqa‐ Derivative-free Bound-constrained Optimizationln-sbplx‐ Subplex variant of Nelder-Meadln-newuoa-bound‐ Derivative-free Bound-constrained Optimization by Iteratively Constructed Quadratic Approximationln-praxis‐ Gradient-free Local Optimization via the Principal-Axis Methodgn-direct-noscal‐ Dividing Rectangles (unscaled)ld-tnewton-precond-restart‐ Preconditioned Truncated Newton with steepest-descent restartinglower= -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-orig-direct-l‐ Dividing Rectangles (original implementation, locally biased)g-mlsl-lds‐ Multi-Level Single-Linkage (low-discrepancy-sequence, require local gradient based optimization and bounds)gn-direct-l-noscal‐ Dividing Rectangles (unscaled, locally biased)gn-isres‐ Improved Stochastic Ranking Evolution Strategyld-tnewton‐ Truncated Newtongn-direct-l-rand‐ Dividing Rectangles (locally biased, randomized)ln-newuoa‐ Derivative-free Unconstrained Optimization by Iteratively Constructed Quadratic Approximationgn-direct-l-rand-noscale‐ Dividing Rectangles (unscaled, locally biased, randomized)gn-orig-direct‐ Dividing Rectangles (original implementation)ld-tnewton-precond‐ Preconditioned Truncated Newtonld-tnewton-restart‐ Truncated Newton with steepest-descent restartinggn-direct‐ Dividing Rectanglesauglag-eq‐ Augmented Lagrangian algorithm with equality constraints onlyln-neldermead‐ Nelder-Mead simplex algorithmln-cobyla‐ Constrained Optimization BY Linear Approximationgn-crs2-lm‐ Controlled Random Search with Local Mutationld-var2‐ Shifted Limited-Memory Variable-Metric, Rank 2ld-var1‐ Shifted Limited-Memory Variable-Metric, Rank 1ld-mma‐ Method of Moving Asymptotesld-lbfgs-nocedal‐ Noneg-mlsl‐ Multi-Level Single-Linkage (require local optimization and bounds)ld-lbfgs‐ Low-storage BFGSgn-direct-l‐ Dividing Rectangles (locally biased)ln-bobyqa‐ Derivative-free Bound-constrained Optimizationln-sbplx‐ Subplex variant of Nelder-Meadln-newuoa-bound‐ Derivative-free Bound-constrained Optimization by Iteratively Constructed Quadratic Approximationauglag‐ Augmented Lagrangian algorithmln-praxis‐ Gradient-free Local Optimization via the Principal-Axis Methodgn-direct-noscal‐ Dividing Rectangles (unscaled)ld-tnewton-precond-restart‐ Preconditioned Truncated Newton with steepest-descent restartingld-slsqp‐ Sequential Least-Squares Quadratic Programmingstep= 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.

**EXAMPLE**

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. The resulting transformation is saved in reg.vf. 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

**COPYRIGHT**

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'.