[Insight-users] Metric values

[motes motes]

I run an image registration method on 3D dicom volumes using the:

template < class TFixedImage, class TMovingImage >
class AdvancedMeanSquaresImageToImageMetric :
   public AdvancedImageToImageMetric< TFixedImage, TMovingImage>
{
public:

from the elastix library and a RegularGradientDescent optimizer
(minimization). When I disable normilization I get values like:


Iteration: 23 Value: 23119
Iteration: 24 Value: 23257.4
Iteration: 25 Value: 20195.5
Iteration: 26 Value: 22521.1
Iteration: 27 Value: 21014.4
Iteration: 28 Value: 25089.2
Iteration: 29 Value: 26273.7
Iteration: 30 Value: 26671
Iteration: 31 Value: 19430.8
Iteration: 32 Value: 27530
Iteration: 33 Value: 25823.7
Iteration: 34 Value: 26696.8
Iteration: 35 Value: 22746.9
Iteration: 36 Value: 18154
Iteration: 37 Value: 25628.6

Is it correct that the above values are the sum of differences in
intensity values for all the pixels in the volume? So if the volumes
contains intensities in the range [0; 255] and one image is black and
the other is white the metric value would be:

255*numberOfPixelsInVolume.