Equivalent Circular Diameter in ImageJ

Hi All,

Does the Analyze function in imageJ have an option for Equivalent Circular Diameter (ECD)? I know it outputs Ferret diameter by selecting shape descriptors, but I haven’t found an option for ECD. I’m trying to find the ECD for some protein particles to use along an equation for protein volume that uses ECD. I haven’t had luck finding an equation to convert from Ferret to ED (if it exists). Anyone encountered this and found a solution?


This plugin seems to implement the required measurements:


Any idea why the plug-in gives values of circularity that are outside the 0-1 range (0=non circular and 1=circular)? I also need the circularity to use in the volume equation, but the plug-in is giving numbers outside this range.


Also, I’m getting different values of area when using the analyze particles and shape filter. I wonder if I’m not using the shape filter correctly or not reading the data correctly. I’m assuming that the same parameters should result in the same values for both Analyze particles and Shape filter (for example Area and Perimeter).

This baffled me too. There are two different definitions for circularity. See the discussion
here: Image J Forum,].

Hi John,

I see what you are saying. Indeed there are different equations used. Did you end up doing something to get the circularity in the range of 0 to 1? I read in your thread that your values ranged up to 12. In my case, the circularity is all over the place. I wonder if I can normalize the values by the largest value. The other thing I was thinking was using the circularity of Analyze particles and ECD from shape filter, but I’d like to use data from the same source. Any suggestions on how to best approach this given you encountered this problem.


I went ahead and calculated the ECD from Analyze particle data and it is in good agreement with the values from the Shape Filter. I think the best way is to use the Analyze Particle data to calculate ECD since the circularity is already in the 0 to 1 range.

The answer was at the bottom of the thread I referenced: If you are using a definition of cirsularity that goes between 0 and 1 divide the output by 4*pi which is about 12… and this will convert the output from IJ to the 0-1 range…


1 Like