[Insight-users] Detect noisy points in curve.

[Luis Ibanez]

Hi SuperZZ


Option A:

You could take the "y" values of all the points and smooth them by
using them as scalar values of  a 1D image, then subtract the original
values from the smoothed ones, and use a threshold on that difference
for locating the outliers.


Option B:

You could use the class

  Insight/Code/Review/
       itkBSplineScatteredDataPointSetToImageFilter.h

to generate a smoothed version of your curve, and again, subtract
the origianl pionts from the smoothed ones in order to detect outliers.

This class is described in the Insight Journal paper:

http://www.insight-journal.org/browse/publication/57
http://hdl.handle.net/1926/140

N-D $C^k$ B-Spline Scattered Data Approximation
Tustison N.J., Gee J.C.
PICSL, University of Pennsylvania



       Regards,


             Luis


-------------------------------
On Sun, Apr 4, 2010 at 8:41 AM, superzz <jxdw_zlf at yahoo.com.cn> wrote:
>
> I have a smooth x-y curve (one x value corresponds one y value) with some
> noisy points. The noisy point's y value differs largely from its neigbhours.
> How can I detect the noisy points?
>
> Thanks
>
> superZZ
> --
> View this message in context: http://old.nabble.com/Detect-noisy-points-in-curve.-tp28132909p28132909.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
>