Particle size distribution of non circular objects



Hello everybody!

I am using SEM images to determine the particle size distribution of particles of approximately 1 um size you can see here.

As my particles are not round at all I tried to use the morphological segmentation plugin of MorpholibJ.
This methode detects the borders of my particles pretty well as you can see in the following picture.

Now my problem is that I use the scale for the ratio of pixels to unit provided by the same picture with annotations and when I use Analyze -> Particle Size it gives me an empty table or just finds 3-4 very small particles while the rest is not detected at all.

I would really appriciate any help as this is very important to me and ImageJ is of course a very complex software :slight_smile:
Thank you very much!


Can you post your Images in a different format they do not show up.


Hey susan-sheehan,

I did upload the images again - they were tiff format and somehow this didnt work well!


Anyone can help me? I looked up a bunch of tutorials and no matter which one i follow i dont get the the areas detected, only very small circular spots although the complete circularity range is selected!

Maybe I am doing something wrong in general?


Hey there,
Not sure why that happens to your images. But i could measure everything when i do a watershed on your segmented image (sample morpholibj). Here is the pipeline i followed.

  1. Covert to a binary image (Process>>Binary>>Make binary),
  2. Invert LUT (Image >> Lookup tables >> Invert LUT)
    3, Apply Watershed (Process >> Binary >> Watershed)
    You can then use the Analyse Particles tool to measure the size of your objects. Here is a screenshot

Hope this helps.


Thank you very much! Now it´s working!


I have the same problem as semuser, so I was wandering whether I can get the size distribution after I get the picture like SEMuser’s second one just using morpholibj ?


To be exact, what can I do to get the particles parameters like area from the second picture that SEMuser gave? I prefer to get the result directly without further processing.


Hello @anliangw,

Have a look at the Analyze Particles tool and the measurements you can set for your image.



why the mean is 0? is it correct?



This result indicates that you measured on your binary image. Just overlay the ROIs on your original image and measure again - then the mean will include the data you expect.

eta :slight_smile:


Thanks for the reply!

How to overlay? sorry I’m new here


it’s ok. @jeep168917

If you have the ROIs saved in the ROI Manager.… all you need to do is select your original image and then toggle the Show All option (in the ROI Manager panel). That will overlay your ROIs on that image then… and you can measure once again.

eta :slight_smile:


I followed the previous steps, and got the image.

I was wondering the average size is correct.

Can you help me out? (is it 1.172um? right?)

Because I want to the the average grain size



The quick answer seems ‘yes’ - according to your snapshot image??

But the best way to get help is to post the original image and then let us know the EXACT steps you took to get those ROIs. Then I can try to re-create your workflow and better see what you need to do to measure what you wish (for example - you want the average size of particles?).

So… could you post an original image (in the original file format in which it was acquired) and then the steps you took to get these ROIs (that’s what I’m assuming the image is above).

Also - in general - we tell users to download and use Fiji is Just ImageJ - it is simply a distribution of ImageJ that comes with a bunch of plugins bundled - ready for you to use out-of-the-box. If you are just getting started, we recommend downloading/using Fiji.



Original image (I can’t upload the tif file, but the scale is 6.7875pixels/um)

Type → 8 bit
Plugins → MorpholibJ → Seqmentation → Morphological Segmentation → Run (Display in Watershed line)

Image → Lookup tables → Invert LUT

Process → Binary → Watershed

Analyze →Analyze Particles (show outlines)

I’d like to know the average grain size and also the grain size distribution.

By the way, I’ll try to use Fiji tomorrow!