Login  Register

Re: problems with the function: analyse particles

Posted by Christian Goosmann on Aug 30, 2012; 8:27am
URL: http://imagej.273.s1.nabble.com/problems-with-the-function-analyse-particles-tp4999891p4999901.html

Hi Kathrin,
hey, you got me surprised. The area fraction shows 100% for each
particle also when I try it with the blob sample from the ImageJ file
menu. It seems also to be dependent on the scaling set in 'Analyze>set
scale...' . Maybe one of the programmers know why the area fraction is
not expressed as fraction of the whole image which would be what I'd
expect. For the time being a workaround for you is to get the total area
of your frame from the image information or by selecting the whole image
(ctl+a) and making a measurement (ctl+m). With this figure you can
calculate each fraction of a particle in an excel  or calc -table.
By the way, have you noticed that the results menu in the results window
can show you a histogram of the distribution in your measurements? Maybe
that is also useful for you.
Good luck
Christian

Kathrin Kolotzek wrote:

> Hey Christian, Hey Larry
>
> thanks for your quick respons.
>
> We are having an alloy which is pressed out of grains. And depending the parameters there are sometimes more and bigger pores and sometimes not. So  I want to determine the  porosity depending the used parameters.
> Furthermore I am interested in the percentage area of the particles  to sort them better out.
>
> Christian I think u mean something like this here: (I shortend it and just listets area and %area) right?
>
>     Area       %area
> 1    12        100
> 2    12        100
> 3    12        100
> 4    446       100
> 5    12        100
>
>
> And here I was wondering why it always says 100% in %area, no matter what size the particle has.
> So do I have this right, in your opinion I should just take the values which are presented in the gap "area" and calculate the %area on my own?
>
> Thanks in regard.
> Kathrin
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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