[Insight-users] Question about OtsuMultipleThresholdsImageFilter

[Luis Ibanez]

Hi Marion,

With the OtsuMultipleThresholdImageFilter you must provide:

1) Number of bins in the Histogram
2) Number of thresholds (e.g. number of expected classes - 1 ).

The actual separation between the thresholds depend on the nature
of your images. More precisely, it depends on the statistical
distribution of intensities.

Is your image a synthetically generated image ?

Maybe a more detailed description of your parameter settings will
help us here.

Could you please post the list of thresholds that you get for
each one of the combinations of input parameters that you are
providing to this filter ?



    Thanks



       Luis



-------------------------
Jaehne Marion wrote:

> Hi @ all!
> 
> I have a question concerning the OtsuMultipleThresholdsImageFilter. In 
> the itk documentation i found that this filter
> creates a labeled image that separates the input image into various 
> classes.
> But what depending on?
> Because in my results i get thresholds like 128, 256, 384, and so on. 
> Depending on the histogramBins. Sometimes i get thresholds that have 128 
> steps or 256 steps.
> With this results it is only like a compression.
> Have i forgotten anything or do i make something wrong?
> 
> Thanks a lot,
> 
> Marion
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
> 
>