Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

My problem is that I have a polygon shape file with 1034 records which are the Greek municipalities. My goal is to compare the shape of every record with the circle. In other words, I want to observe the deviation of the polygons from the perfect shape-circle. Thank you!

share|improve this question
up vote 2 down vote accepted

If you have access to ArcMap 10, then Bounding Containers has an option create minimum area bounding circles around features. That would get you started to at least compare the actual area to that of a circle bounding its extent.

share|improve this answer

Standard methods of comparing polygons to circles exploit some of the defining characteristics of circles. One is that of all plane figures of a given area, the circle has the smallest perimeter. To compare an area (measured in square units) to a perimeter (measured in linear units), we deduce the circle's diameter d from its area, A:

d = 2 Sqrt(A/pi).

Its perimeter is

L = d * pi.

Both d and L are in linear units. We therefore compare 2*Sqrt(A/pi) to L/pi where A is the area of an arbitrary polygon and L is the length of its perimeter. The ratio is

r = L/pi / (2*Sqrt(A/pi)) = L/Sqrt(A) / (2*Sqrt(pi)) = 0.282095 * L/Sqrt(A).

By construction, this ratio is always 1 or greater and equals 1 if and only if the polygon is a circle. Therefore you can interpret (r - 1) as a measure of deviation from a perfect circular shape.

As an example, consider a square of side s. Its area is A = s^2 and its perimeter is L = 4 s. We find

r = 0.282095 * 4s / Sqrt(s^2) = 0.282095 * 4 = 1.12838.

This is 0.12838 greater than 1. Note how this measure is independent of the size (s), position, or orientation of the polygon: it depends solely on its shape.

This idea is closely related to that of the tortuosity of a curve.

share|improve this answer

Just for completeness I have managed to find my code for calculating shape metrics of polygons and moved it to

It provides implementations of Blair & Bliss, Lee & Salle, Ritter, Boyce & Clark, Miller, Gibbs and Zusne's methods. See for fuller details of the methods.

share|improve this answer
+1 Good reference. The full article is available at (Beware: the descriptions of the area:perimeter metrics are incorrect.) The most striking element of the article is its characterization of "compactness" as the mean distance between all pairs of points within the shape (relative to a given probability distribution of those points). This is a 4D integral which can be done efficiently (for uniform distributions) via a triangulation of the shape. BTW, a disk of area A has "compactness" of 0.51083 * Sqrt(A). – whuber Jun 17 '11 at 18:47

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.