Re: Summing areas within larger areas

Posted by Stephan Saalfeld on Aug 12, 2008; 7:44pm
URL: http://imagej.273.s1.nabble.com/Summing-areas-within-larger-areas-tp3695399p3695401.html

Sorry for that misplaced answer.  The ImageJ-mailing list rejects signed
mails such that I had to resend my answer to Luciano Marpegan.
Obviously I did something wrong during this attempt.

Best regards,
Stephan
       

On Tue, 2008-08-12 at 20:59 +0200, Stephan Saalfeld wrote:

> Hi Luciano,
>        
> to some extent it depends on how big your largest factor actually is.
> As long as it is smaller than 255, you are fine with 8bit after
> dividing
> by 1023.  Then you could save the images as GIF.  Typically, PNG is
> the
> best choice for 8bit greyscale or 24bit rgb images with large
> homogeneous regions, but I noticed some unexpected contrast changes
> during my last attempts with ImageJ's PNG export.
>
> Best regards,
> Stephan
>
>
> On Tue, 2008-08-12 at 11:27 -0600, Darren Johnson wrote:
> > Hello all,
> >
> > I have only been using ImageJ for a couple of weeks but have found
> it very useful in my research. I am looking for a way to sum smaller
> areas that reside in larger areas. For example, I have an image
> showing several hundred grains that I have done particle analysis on
> and therefore have the outlines (and areas) of each of the individual
> grains. I also have an image that shows iron bearing minerals within
> these grains. I have done a separate particle analysis on the iron
> bearing image and also have the outlines for them. I was wondering if
> there is a way to use the separate outline images to sum all of the
> small areas (iron bearing minerals) that lie within the boundaries of
> larger areas (i.e. each individual grain). An added bonus would be if
> I could retain the original numbers associated with individual grain
> image, although this is not as important. It is virtually impossible
> to manually do this because some of the grains may have 40 to 50 Fe
> bearing minerals (small areas) so the numbers overlap and are
> difficult to distinguish.
> >
> >
> > Thanks,
> >  
> > Darren Johnson
> > Washington University in St. Louis