A suggestion: Make a scatterplot available, plotting the pixel-by-pixel values of any two relevant quantities, and make it possible to define a polygon mask on that scatterplot. That often allows separation of chemically-distinct areas. Sincerely, Matthew Marcus On 11/21/2019 12:26 PM, Matt Newville wrote:

HI Troy,

Sorry for the late reply -- I think this sort of slipped off my radar screen.

On Sat, Nov 16, 2019 at 2:09 PM E. Rasbury mailto:troy.rasbury@stonybrook.edu> wrote:

I have used larch some for XRF maps. I like the feature for defining ROI's and then looking at the statistics of the data. I have some samples that are an intimate mix of two or more phases and I would like to extract pixels based on a criteria- like high Sr or low Fe, etc, from across a whole map region. I wonder if this feature is available, or would be easy to create on larch?

Like you say, this is not currently available in the Mapviewer GUI.  But it's a good suggestion and might be possible to do with a small amount of code.

Would it work for you to say "Make an Area with all pixels where Sr Counts is above some value"?  I think that would be easy to generalize from "Sr Counts" to "Ratio of ROI Counts"  (or "analyzed concentration" once I get that done - a work in progress).  If that's right, I think it would be "not hard to add".

--Matt

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Hi Matthew,



On Thu, Nov 21, 2019 at 2:33 PM Matthew Marcus  wrote:

> A suggestion:  Make a scatterplot available, plotting the pixel-by-pixel
> values of any two relevant quantities, and make it
> possible to define a polygon mask on that scatterplot.  That often allows
> separation of chemically-distinct areas.
>
>
Thanks!   The Mapviewer has an easy way to make a "Corrrelation plot" for
two ROI maps.  This includes a scatter plot of ROI1 + ROI2, the two
individual maps of ROI1 and ROI2 and the joint map in two complementary
colors.  Users can draw a polygon mask on the scatterplot to highlight
where different correlations appear.  But right now, these are not saved as
"areas" (masks that can then be used for XRF analysis or the sorts of
statistics Troy was referring to).  But that "scatterplot selection of
pixels to area" would be pretty easy to add, and it makes sense to be able
to save those selections.   So, I will add that.  Thanks for the suggestion!


Sincerely,
>      Matthew Marcus
>
> On 11/21/2019 12:26 PM, Matt Newville wrote:
> > HI Troy,
> >
> > Sorry for the late reply -- I think this sort of slipped off my radar
> screen.
> >
> >
> > On Sat, Nov 16, 2019 at 2:09 PM E. Rasbury  > wrote:
> >
> >     I have used larch some for XRF maps. I like the feature for defining
> ROI's and then looking at the statistics of the data. I have some samples
> that are an intimate mix of two or more phases and I would like to extract
> pixels based on a criteria- like high Sr or low Fe, etc, from across a
> whole map region. I wonder if this feature is available, or would be easy
> to create on larch?
> >
> >
> > Like you say, this is not currently available in the Mapviewer GUI.  But
> it's a good suggestion and might be possible to do with a small amount of
> code.
> >
> > Would it work for you to say "Make an Area with all pixels where Sr
> Counts is above some value"?  I think that would be easy to generalize from
> "Sr Counts" to
> > "Ratio of ROI Counts"  (or "analyzed concentration" once I get that done
> - a work in progress).  If that's right, I think it would be "not hard to
> add".
> >
> > --Matt
> >
> >
> > _______________________________________________
> > Ifeffit mailing list
> > Ifeffit@millenia.cars.aps.anl.gov
> > http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> > Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit
> >
> _______________________________________________
> Ifeffit mailing list
> Ifeffit@millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
> Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit
>


-- 
--Matt Newville  630-252-0431