[Insight-users] Approximate segmentation of very high intensity region.

[wanlin]

HI, xi
    May be you could try a similar way like this
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1352428&tag=1

1. you always define 2 classes. the background and foreground (only for
nonzero pixels)
2. compute mean value for each class.
3. perform k Means classification.
4. remove background (set to 0). test whether foreground satisfied your
condition(like variance < threshold)
5. if 4 not satisfied, go to 2,3,4.


regards

wanlin


On Tue, Oct 26, 2010 at 10:04 AM, Xi Liang <xliang.cad at gmail.com> wrote:

>
> I tried KMeans by splitting the subtraction images into two/three/four
> clusters.
>
> The initial means is set to be (0, MaxIntensity) where the interval is
> equal
> to MaxIntensity/#Clusters. I tested k=2,3,4,5 means method,  3-means method
> shows best segmenation in my case by using the cluster with the highest
> intensity mean as tumor mask.
>
> Kind regards,
>
> Xi
> --
> View this message in context:
> http://old.nabble.com/Approximate-segmentation-of-very-high-intensity-region.-tp30052326p30052691.html
> Sent from the ITK - Users mailing list archive at Nabble.com.
>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> http://www.kitware.com/products/protraining.html
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20101026/cf16570a/attachment.htm>