[Insight-users] itkKmeansModelEstimator

[Luis Ibanez]

Hi Tom,

You could use the BinaryThresholdImageFilter after the classifier,
and select each component by using its label number.


    Regards,


        Luis


------------
Tom Nash wrote:
> Hi All
> 
> Taking both Luis's and Karthik's advice, i have redirected the output of 
> the ImageKmeansModelEstimator to the input of the ImageClassifierBase 
> Classifier and used the membershipFunctions produced in the estimator 
> and successfully got the result after classification. However I have 
> basically 4 classes and want 4 separate output files for each one of the 
> classes. I am basically trying to segment the classified image. However 
> in ImageClassifierBase  I can get only 4 classes together. I can do an 
> iterative technique and classify the classified image into 3 classes and 
> then into 2 classes and so on and use SubtractImageFilter to get the 
> components. However I want to retain the spatial information. Is there 
> anyway to separate the 4 classes from the classified image? I am unable 
> to find such a filter. Please advise.
> 
> Thanks a lot for everything.
> 
> Sincerely
> Tom
> 
> 
> 
>  > Date: Thu, 16 Oct 2008 12:01:04 -0400
>  > From: karthik.krishnan at kitware.com
>  > To: tomnash1 at hotmail.com; insight-users at itk.org
>  > Subject: Re: itkKmeansModelEstimator
>  >
>  > Tom:
>  >
>  > You might be looking from ScalarImageKMeansFilter, which will
>  > generate the KMeans followed by a classification based on a
>  > Euclidean minumum distance rule to generate a label map
>  > segmentation of your image.
>  >
>  > It internally use the itkKMeansModelEstimator to generate the
>  > KMeans. And then uses a itk::EuclideanDistance in conjuction
>  > with a itk::MinimumDistance rule to generate the label map
>  > based on those means.
>  >
>  > hope this helps
>  >
>  > On Tue, Oct 14, 2008 at 10:49 PM, Tom Nash <tomnash1 at hotmail.com> wrote:
>  > > Hi Karthik
>  > >
>  > > I am using itkKmeansModelEstimator for k-means segmentation. 
> However i see
>  > > no function to view the output image after classification. The older
>  > > ImageKmeansImageFilter had a kmeansFilter->GetOutput() function, 
> but I see
>  > > no analogous function in the itkKmeansModelEstimator. How do I get the
>  > > output? I want to parse it into the writer object. I am using nifti 
> images.
>  > >
>  > > Please help
>  > >
>  > > Sincerely
>  > > Tom
>  > >
>  > > ________________________________
>  > > Hottest news and in-depth analysis that goes beyond the headlines. 
> Only on
>  > > MSN News Check it out!
>  >
>  >
>  >
>  > --
>  > Karthik Krishnan
>  > R&D Engineer,
>  > Kitware Inc.
>  > Ph: 518 371 3971 x119
>  > Fax: 518 371 3971
> 
> ------------------------------------------------------------------------
> Watch useful tips on recipes, fitness, yoga and fashion only on MSN 
> videos. Try it! <http://video.msn.com/?mkt=en-in>