Hi, I am trying to quantify cell shape based on roundness vs elongation/spindle-like shape. I was wondering what the difference was between the circularity function and the roundness function within the shape descriptors measurement in Image J?
I was thinking to use circularity, but some of my cells are irregularly shaped (neither circular nor elongated), so I wasn't sure if this would be appropriate.
I did a quick test, making an image with a perfectly round circle and one that is twice as tall as it is wide. See the image. Roundness gives a value direlectly relative to the aspect ratio, circularity something more complex. Hope this helps a little.
See the user guide. A link is here to the subsection where exact formulas are given for circularity and roundness measurements in ImageJ.
https://imagej.nih.gov/ij/docs/guide/146-30.html#toc-Subsection-30.7
Thank you. I saw the user guide, but I am having trouble distinguishing the difference between circularity and roundness. They seem to measure the same thing according to the manual.
I strongly encourage you to take a look at our new software MIPAR (http://MIPAR.us). You could quickly segment your features and color them according to eccentricity (a measure of "circularity") and convex area/area ratio (a measure of "roundness"/convexity - circle = low, star = high). With an example image, I'd be happy to make a recipe for you to auto-find your features, and even make a screen recording showing the shape measurements.
Here's an example image. It is within a 3D scaffold (so several layers of cells)
I did a quick test, making an image with a perfectly round circle and one that is twice as tall as it is wide. See the image. Roundness gives a value direlectly relative to the aspect ratio, circularity something more complex. Hope this helps a little.
Here's how I think of it (and, please, correct me if I am mistaken here):
In ImageJ, circularity is 4pi(Area)/(Perimeter^2), while roundness is 4A/(pi*Major Axis^2).
Consider a rectangle and an ellipse of similar dimensions. Because they have similar dimensions and thus similar areas and major axes, their roundness should be similar. However, because circularity depends on the perimeter of the object, a less smooth object (the rectangle) will have a lower circularity than a more smooth object of a similar area (the ellipse).
- Badges
- Science topic
- Similar topics
- Engineering
- Electrical Engineering
- Signal Processing
- Image Processing