Plugin type: 3dimage/cost

3D image similarity kernels evaluate the according similarity measure between two images. These kernels may be used standalone, like e.g. in linear registration, or will be called from generalized image similarity cost plug-ins that also take care of transforming and scaling the images during the image registration process.

Plugins:

lncc mi ncc ngf ssd ssd-automask

lncc

local normalized cross correlation with masking support.. Supported parameters are:

NameTypeDefaultDescription
wuint in [1, 256]5half width of the window used for evaluating the localized cross correlation

mi

Spline parzen based mutual information.. Supported parameters are:

NameTypeDefaultDescription
cutfloat in [0, 40]0Percentage of pixels to cut at high and low intensities to remove outliers
mbinsuint in [1, 256]64Number of histogram bins used for the moving image
mkernelfactory[bspline:d=3]Spline kernel for moving image parzen hinstogram. For supported plug-ins see Plugin type: 1d/splinekernel
rbinsuint in [1, 256]64Number of histogram bins used for the reference image
rkernelfactory[bspline:d=0]Spline kernel for reference image parzen hinstogram. For supported plug-ins see Plugin type: 1d/splinekernel

ncc

normalized cross correlation.. (This plug-in doesn't take 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:

NameTypeDefaultDescription
evaldictds
plugin subtype (sq, ds,dot,cross)
ds:square of scaled difference
dot:scalar product kernel
cross:cross product kernel

ssd

3D image cost: sum of squared differences. Supported parameters are:

NameTypeDefaultDescription
autothreshfloat in [0, 1000]0Use automatic masking of the moving image by only takeing intensity values into accound that are larger than the given threshold
normbool0Set 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:

NameTypeDefaultDescription
rthreshdouble0Threshold intensity value for reference image
sthreshdouble0Threshold intensity value for source image

Plugin consumers:

mia-3dcost mia-3dcost-translatedgrad mia-3dforce mia-3dmany2one-nonrigid mia-3dnonrigidreg mia-3dnonrigidreg-alt mia-3dprealign-nonrigid mia-3drigidreg mia-3dserial-nonrigid image:3dimage/fullcost