ITK registration optimizer scaling, offset and step length

The recommendation for the scaling of translation parametersversus rotation parameter is to use a factor proportional tothe diagonal length of the image.For your case the, you have 100 pixels with 1 mm / pixel,therefore the physical extent of your image is        100mm  X  100mm  X 100mmThe diagonal the image bounding box is          sqrt(3) * 100 mmwhich is about              173.2and extra factor of 10X is usually useful, so you shouldprobably try a factor of     1.0 / ( 10 x 173.2 )  =  1.0 / 1732.0You could use this same factor for the three componentsof the translation or you could estimate independentfactor for each component in the way it is done in theVolView plugin.Note that this factors are not expected to be computedprecisely. Their purpose is simply to bring the rotationaland translational parameters to a similar numerical scale.By default, they are quite disproportionate since rotationare in radians, therefore in a range about -1:1, whiletranslations are in millimeters, and for an image of 100mmyou probably can expect translations as large as 50mm.The difference between Offset and Translation is relevantonly for the "Centered" transforms.  For example, for thecase of the CenteredAffineTransform the full transformationis given by           P' = R x ( P - C ) + ( C + T )where   C  is the Center of rotation   P  is the point to be transformed   P' is the transformed point   T  is the translation.This equation can be rewritten as        P' =   R x P     +     [ C + T - R X C ]and we call Offset the expression      Offset = [ C + T - R X C ]so the transformation is       P'  =  R x P  + OffsetSo, the relationship between Translation and offset is      Offset =  [ I - R ] x C    +    TIn practice, if you are using Centered transform, you shouldnot care about the Offset. Instead make sure that you provideappropriate values for the Center of Rotation and the Translation.The Offset is computed from these two and the rotation matrix.Note that the step length is also a critical value.There is no magic recipe for selecting one. You probablywant to start experimenting with a small value (e.g. 0.01)and plot the metric evaluations during the registrationprocess.  If you observe that the metric values are fairlymonotonic, that means that you can safely increment thestep length. Such an increment has the advantage of reducingthe time required to reach an extrema of the cost function(the image metric in this case).   You could restart theregistration with larger values of the step length, as longas you don't observe a noisy and/or erratic behavior on theMetric values.