Lower circularity limit

[Kristoffer Persson]

Hi,
Anyone that knows the lower area limit for a particle before it always
gives circularity = 1?


--
Kristoffer

[Angel Cruz-3]

Hi Kristoffer,

The circularity related to the form, therefore with a particle of 1 pixel
for 2 pixel gives 1, but on a particle of 3x1 already it meets differently.
Providing that the object is more or less a proportional uniform as a square
or a perfect circle is 1. Not this so associated with the area since the
circularity is unvariant to the scale, the circularity is related to the
form of the object.


2007/3/21, Kristoffer Persson <[hidden email]>:
>
> Hi,
> Anyone that knows the lower area limit for a particle before it always
> gives circularity = 1?
>
>
> --
> Kristoffer
>



--
Cordial Saludo,


ANGEL ALFONSO CRUZ ROA
System Engineer -Universidad de los Llanos
Young Researcher - Colciencias
Instituto Nacional de Salud
Laboratorio de Microscopia y Analisis de Imagenes
Bogota D.C.,  COLOMBIA
Tel. + 57 1 220 7700 - Ext. 453
Movil: 300 - 586 0237
E-mail:  [hidden email],
E-mail:  [hidden email]
Messenger: [hidden email]
Skype: tenshi_pdi

"In Nomine Gaia Missit me Dominus"

[Gabriel Landini]

On Wednesday 21 March 2007 17:49:08 Angel Cruz wrote:
> The circularity related to the form, therefore with a particle of 1 pixel
> for 2 pixel gives 1, but on a particle of 3x1 already it meets differently.
> Providing that the object is more or less a proportional uniform as a
> square or a perfect circle is 1.

> Not this so associated with the area since
> the circularity is unvariant to the scale, the circularity is related to
> the form of the object.

Only in theory, but certainly not in practice when dealing with pixellated
binary objects (specially small ones).
You will have to do a test with increasingly larger (or smaller) circles to
see how much departure from the ideal value one gets.
This depends on the way area and perimeter are defined; there are various ways
to measure area and perimeter so the circularity can vary according to which
method is used (in general they will be closely correlated, but the values
will not be the same).
For example, circularity with the built in particle analyzer is different from
circularity with the Particles_8 plugin.

This is explained in the Technical FAQ 5 here:
http://imagejdocu.tudor.lu/imagej-documentation-wiki/faq

Cheers,

Gabriel