[Paraview] Interpretation of histogram of deformation field magnitude?

[Utkarsh Ayachit]

On the properties panel for the Histogram filter, you choose the array
to generate histogram from. Currently the panel doesn't not support
selecting the magnitude of a vector array (only 1 component). In your
case, by default the 0th component of the GlyphVector is chosen, which
can be negative and hence the plot.

To generate a histogram for the magnitude of the vector, apply another
calculator with the expression:

mag(GlyphVector)

and then generate a histogram using the "Result" array.

Utkarsh

On Tue, Nov 3, 2009 at 8:15 PM, motes motes <mort.motes at gmail.com> wrote:
> I have loaded a deformation field (.mhd file) in paraview made with itk:
>
> 1) open .mhd file
> 2) connect the calculate and use the below expression:
>
>  MetaImage_X*iHat+MetaImage_Y*jHat+0*kHat
>
> 3) Add glyph filter.
>
> The result looks like this:
>
> http://37133.vs.webtropia.com/apache2-default/test/deform.png
>
>
> But now I would like to create a histogram over the magnitude of the
> deformation field. I have therefore selected: Filters->Data
> Analysis->Histogram.
>
> In the display tab I have selected GlypVector_total (Magnitude) and
> under "User Data Array" I have selected MetaImage. The result looks
> like this:
>
> http://37133.vs.webtropia.com/apache2-default/test/hist.png
>
> But how should it be interpretated? I assume that the y-axis is the
> pixel count and the x-axis is the magnitude size. But should the
> magnitudes (Norm) not always be positive?
>
> And as can be seen from the deformation field its only a small part of
> the image that is deformed. I would therefore expect a large number of
> pixels to have 0 magnitude, but that is not the case.
>
> Any input are welcome!
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>