<div dir="ltr">thx, I will try vtkImageHistogramStatistics, right now I don't want itk as dependence.<br></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Mar 4, 2013 at 10:03 AM, David Gobbi <span dir="ltr"><<a href="mailto:david.gobbi@gmail.com" target="_blank">david.gobbi@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Yes, there are statistical methods that you can use to automatically<br>
find the best thresholds between different tissue types in an image.<br>
For example, K-means clustering. ITK has several of these clustering<br>
algorithms. Or you can use vtkImageHistogramStatistics and set your<br>
thresholds based on percentiles, if you know the approximate<br>
proportions of the various tissue types that make up the image.<br>
<span class="HOEnZb"><font color="#888888"><br>
- David<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
On Mon, Mar 4, 2013 at 7:43 AM, José M. Rodriguez Bacallao<br>
<<a href="mailto:jmrbcu@gmail.com">jmrbcu@gmail.com</a>> wrote:<br>
> there is an automated way to find those values?<br>
><br>
><br>
> On Thu, Jun 2, 2011 at 1:10 PM, David Gobbi <<a href="mailto:david.gobbi@gmail.com">david.gobbi@gmail.com</a>> wrote:<br>
>><br>
>> Hi Dora,<br>
>><br>
>> The isovalues and color transfer functions in the medical examples are all<br>
>> hard-coded for the intensity values of the example data. You will have to<br>
>> find out what values to use for your own data.<br>
>><br>
>> - David<br>
>><br>
>><br>
>> On Wed, Jun 1, 2011 at 7:05 AM, Dora Szasz <<a href="mailto:dora.szasz@yahoo.com">dora.szasz@yahoo.com</a>> wrote:<br>
>> > Hello all,<br>
>> > I have tried to use Medical examples with another set of Dicom images,<br>
>> > but<br>
>> > not works ok. Instead of seeing for example a skin, I see a strange<br>
>> > volume.<br>
>> > Can anyone tell me why and what should I change?<br>
>> > Thank you!<br>
>> > Dora<br>
</div></div></blockquote></div><br></div>