<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:times new roman, new york, times, serif;font-size:12pt"><DIV>Hi, Luis:</DIV>
<DIV> </DIV>
<DIV>I have look the following code in Insight/Examples/Statistics/ ScalarImageMarkovRandomField1.cxx<BR></DIV>
<DIV> </DIV>
<DIV> </DIV><FONT face=Courier size=1>
<P align=left>typedef itk::Statistics::DistanceToCentroidMembershipFunction<</P>
<P align=left>ArrayPixelType ></P>
<P align=left>MembershipFunctionType;</P>
<P align=left>typedef MembershipFunctionType::Pointer MembershipFunctionPointer;</P>
<P align=left>double meanDistance = 0;</P>
<P align=left>vnl_vector<double> centroid(1);</P>
<P align=left>for( unsigned int i=0; i < numberOfClasses; i++ )</P>
<P align=left>{</P>
<P align=left>MembershipFunctionPointer membershipFunction =</P>
<P align=left>MembershipFunctionType::New();</P>
<P align=left>centroid[0] = atof( argv[i+numberOfArgumentsBeforeMeans] );</P>
<P align=left>membershipFunction->SetCentroid( centroid );</P>
<P align=left>classifier->AddMembershipFunction( membershipFunction );</P>
<P align=left>meanDistance += static_cast< double > (centroid[0]);</P>
<P align=left> </P>
<P align=left>}</P>
<P align=left> </P>
<P align=left>Certainly, the program need user to input the centroid which I guess is the mean intensity value of each cluster? Am I right?</P>
<P align=left> </P>
<P align=left>My qustion is how accurate the program relies on the input mean. Since I already have reference labeled image, it is not a big deal to calculate the mean.</P>
<P></FONT> </P>
<P>Can you furthe tell me whehter I can get the similar effect by using itk EM clustering and itkMRFImageFilter with the method by Zhang et al. (many again is a naive question)</P>
<P> </P><FONT face=Times-Roman size=1><FONT face=Times-Roman size=1>
<P align=left></FONT><FONT face=Times-Roman size=1>Zhang, Y., M. Brady, and S. Smith. Segmentation of brain MR images through a hidden Markov random field model and the</P>
<P align=left>expectation-maximization. </FONT><I><FONT face=Times-Italic size=1>IEEE Trans. Med. Imaging </I></FONT><FONT face=Times-Roman size=1>20:45–57, 2001</FONT></P>
<P align=left><FONT face=Times-Roman size=1></FONT> </P>
<P align=left><FONT face=Times-Roman size=1>Thanks</FONT></P>
<P align=left><FONT face=Times-Roman size=1></FONT> </P>
<P align=left><FONT face=Times-Roman size=1>Baoyun.</P>
<P align=left></FONT></FONT><BR> </P>
<DIV style="FONT-SIZE: 12pt; FONT-FAMILY: times new roman, new york, times, serif"><BR>
<DIV style="FONT-SIZE: 13px; FONT-FAMILY: arial, helvetica, sans-serif"><FONT face=Tahoma size=2>
<HR SIZE=1>
<B><SPAN style="FONT-WEIGHT: bold">From:</SPAN></B> Luis Ibanez <luis.ibanez@kitware.com><BR><B><SPAN style="FONT-WEIGHT: bold">To:</SPAN></B> Baoyun Li <baoyun_li123@yahoo..com><BR><B><SPAN style="FONT-WEIGHT: bold">Cc:</SPAN></B> insight-users@itk.org<BR><B><SPAN style="FONT-WEIGHT: bold">Sent:</SPAN></B> Saturday, March 21, 2009 12:35:48 PM<BR><B><SPAN style="FONT-WEIGHT: bold">Subject:</SPAN></B> Re: can itkMRFImageFilter discard the intensity information<BR></FONT><BR><BR>Hi Baoyun,<BR><BR><BR>Please take a look at the example:<BR><BR> Insight/Examples/Statistics/<BR> ScalarImageMarkovRandomField1.cxx<BR><BR>You may find it to be a useful guide on how to use the MRF filter.<BR><BR>I'm not sure what you mean by "using the intensity information".<BR>Could you please explain this in more detail ?<BR><BR>A piece of source code will be great....<BR><BR><BR> Thanks<BR><BR><BR>
Luis<BR><BR><BR><BR>------------------<BR>Baoyun Li wrote:<BR>> Dear All:<BR>> After going through the documents for itkMRFImageFilter, I have some doubts about using the intensity information.<BR>> I understand the filter using Gassian model to relabel the segmeantion image. But why?<BR>> In otherwords, if I have already got ok segmeantion result based EM Gaussian Mixuture model, thus I only need hMRF to take care of the continous of the labled segmetation. Seem reclassify the image based on Gassuian model is redundent or make the segmentation worse if I blieved my initial segmentation.<BR>> If there anyway to ignore using the intensity information.<BR>> Or I am fundmentally wrong, say that using intensity information certainly can improve the performance.<BR>> Can somebody teach me?<BR>> Best regards<BR>> Baoyun<BR>> <BR></DIV></DIV></div><br>
</body></html>