mia-3dprealign-nonrigid
{-i io
} {-o io
} [
options
...]
This program runs the non-rigid registration of an image series by first registering an already aligned subset of the images to one reference, and then by registering the remaining images by using synthetic references. This is a 3D version of
input images following the naming pattern nameXXXX.ext. For supported file types see Plugin type: 3dimage/io
file name base for registered files given as C-format string. For supported file types see Plugin type: 3dimage/io
Save synthetic references to files refXXXX.v
Skip images at the begin of the series
Skip images at the beginning+skip of the series when searching for high contrats image
Skip images at the end of the series when searching for high contrats image
maximum number of candidates for global reference image
Const function to use for the analysis of the series. For supported plug-ins see Plugin type: 3dimage/fullcost
save reference index number to this file
save reference index number to this file
Maximum delta between two elements of the prealigned subset
Optimizer used for minimization. For supported plug-ins see Plugin type: minimizer/singlecost
multi-resolution levels
transformation type. For supported plug-ins see Plugin type: 3dimage/transform
Cost function for registration during the subset registration. For supported plug-ins see Plugin type: 3dimage/fullcost
Cost function for registration during the final registration. For supported plug-ins see Plugin type: 3dimage/fullcost
verbosity of output, print messages of given level and higher priorities. Supported priorities starting at lowest level are:
debug: | Debug output |
message: | Normal messages |
info: | Low level messages |
error: | Report errors |
fatal: | Report only fatal errors |
trace: | Function call trace |
warning: | Warnings |
fail: | Report test failures |
print copyright information
print this help
print a short help
print the version number and exit
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).
Register the image series given by images imageXXXX.v by optimizing a spline based transformation with a coefficient rate of 16 pixel, skipping two images at the beginning and using normalized gradient fields as initial cost measure and SSD as final measure. Penalize the transformation by using divcurl with aweight of 2.0. As optimizer an nlopt based newton method is used.
mia-3dprealign-nonrigid mia-3dprealign-nonrigid -i imageXXXX.v -o registered -t vista -k 2-F spline:rate=16,penalty=[divcurl:weight=2] -1 image:cost=[ngf:eval=ds] -2 image:cost=ssd -O nlopt:opt=ld-var1,xtola=0.001,ftolr=0.001,maxiter=300
Gert Wollny