[Insight-users] Questions about the ScalarImageMarkovRandomField1.cxx

[Jimmy Wong]

Dear all,

I've read through the example in 
\Insight\Exmaples\Statistics\ScalarImageMarkovRandomField1.cxx and have some 
questions about this example.

First, I note that we need the specify the mean value for each class. So, 
here comes a question, are they initialization or are they some important 
parameters for this example? I change the mean values to arbitary values and 
get poor MRF results.

For this example, the alpha(parameters set for the MRF) is assumed to be 
known, right? That means we don't need to update the alpha for the MRF, what 
we need to do is to estimate the parameters for the data set. Am I wrong? 
So, this is not the hidden MRF.

May I know where I can find a tutorial for this example or a diagram for it?

Thank you if you can help.

Best regards,

Zhimin