Re: tortuosity

Posted by John Dunsmuir on Jul 19, 2013; 1:56pm
URL: http://imagej.273.s1.nabble.com/tortuosity-tp5004032p5004036.html

I would suggest a different approach.  Have a look at "Practical Methods for Measuring the Tortuosity of Porous Materials from Binary or Gray-Tone Tomographic Reconstructions", C.J. Gommes et.al.. AICHE Journal Vol. 55, No. 8.  The iteration count for a 6 connected flooding kernel is used as an estimator for geodesic distance in the pore space.  Dividing the resulting image by the Euclidean distance produces a tortuosity image.  If you are flooding from the top slice to the bottom slice the Euclidean distance is simply the slice number.

If the image is sufficiently large and is a random network then the tortuosities should converge to a solution as distance increases from the seed point or seed plane (usually an image face).  This process tends to overestimate the tortuosity but is still a useful metric.

A plugin that does 6 connected 3D flood filling will work but you will need to modify it to report the iteration count.

Regards,
John D.

prasanthriver wrote

Hi users,

  I am trying to find the tortuosity of a material(3D) In x,y,z directions.
1)As a first step I skeletonized the pores.
2)I used analyze skeleton in imagej to get the vertexes and the edges.

How do i identify the shortest path from this ? how do i draw all the possible paths in a graph and find the tortuosity  distribution in all three directions?