[Insight-users] standard deviation parameter in MutualInformationmetric

[Laurent Paul]

Dear all,

I use the Mutual Information for a while and it works perfectly.
I tried to apply it onto non normalized images and sometimes, I get the 
following error message:
"Standard deviation is too small"

I saw that a gaussian kernel is used to estimate the probability density 
function and that the default value for its standard deviation is 0.4.
Is there a way to estimate the optimal value for this parameter? Is it 
dependant to the input images grey values ?
Thanks for your help.
Laurent.