Tweaking the particle analyzer

[Thomas Eschner]

[Thomas Eschner]

...apparently, the text disappeared automagically. Here it is:
 

Dear all,
 
I am struggling with a task, that first seamed quite easy to me, but now I run out of ideas...
 
I want to determine the percentage of dark particles with respect to the white area (including the area of contained dark particles) - see attached image - and to add a single line entry to the summary window, very similar to what is done by the particle analyzer.
 
"Remove on edge" successfully removes the upper large "particle", but the data still relates to the whole window.
 
Is anybody aware of an approach that does the trick without going for programming? Any hint is welcome...
 
Thanks in advance.
 
Thomas

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Michael Schmid]

Hi Thomas,

here is a simple way to get the fraction of particle area in a region:

Start with a binary image (particles=255, everything else=0).

Use the Wand Tool to select the dark 'non-particle' at the top.
Then, Edit>Selection>Make Inverse to select your region of interest.

If you simply press Measure, the 'Mean' column will give you 255 times the area fraction of the particles in the region of interest. Thus, simply divide the 'Mean' value by 255.  In your case, it's 0.47766/255 = 0.00187 or 0.187%.

Michael
________________________________________________________________
On Jul 11, 2013, at 21:38, Thomas Eschner wrote:

> ...apparently, the text disappeared automagically. Here it is:
>  
>
> Dear all,
>  
> I am struggling with a task, that first seamed quite easy to me, but now I run out of ideas...
>  
> I want to determine the percentage of dark particles with respect to the white area (including the area of contained dark particles) - see attached image - and to add a single line entry to the summary window, very similar to what is done by the particle analyzer.
>  
> "Remove on edge" successfully removes the upper large "particle", but the data still relates to the whole window.
>  
> Is anybody aware of an approach that does the trick without going for programming? Any hint is welcome...
>  
> Thanks in advance.
>  
> Thomas

... [show rest of quote]

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Thomas Eschner]

Hi Michael,

thank you.

In my mail I didn't tell that I want to do this without manual interaction as I have to do this on a large number of images.

Meanwhile I managed to solve my problem after I gave up the idea of doing it in one step. With setting backgrounds properly and closing holes I ended up with the percentage of the large area. I a second run I determined the relative amount of particles. From these numbers I calculated what's needed and this ist an approach that also works for file sets.

Kind regards,
Thomas
 
 

Gesendet: Montag, 15. Juli 2013 um 16:35 Uhr
Von: "Michael Schmid" <[hidden email]>
An: [hidden email]
Betreff: Re: Tweaking the particle analyzer
Hi Thomas,

here is a simple way to get the fraction of particle area in a region:

Start with a binary image (particles=255, everything else=0).

Use the Wand Tool to select the dark 'non-particle' at the top.
Then, Edit>Selection>Make Inverse to select your region of interest.

If you simply press Measure, the 'Mean' column will give you 255 times the area fraction of the particles in the region of interest. Thus, simply divide the 'Mean' value by 255. In your case, it's 0.47766/255 = 0.00187 or 0.187%.

Michael
________________________________________________________________
On Jul 11, 2013, at 21:38, Thomas Eschner wrote:

> ...apparently, the text disappeared automagically. Here it is:
>
>
> Dear all,
>
> I am struggling with a task, that first seamed quite easy to me, but now I run out of ideas...
>
> I want to determine the percentage of dark particles with respect to the white area (including the area of contained dark particles) - see attached image - and to add a single line entry to the summary window, very similar to what is done by the particle analyzer.
>
> "Remove on edge" successfully removes the upper large "particle", but the data still relates to the whole window.
>
> Is anybody aware of an approach that does the trick without going for programming? Any hint is welcome...
>
> Thanks in advance.
>
> Thomas

... [show rest of quote]

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html