[Insight-users] How to generate voronoi mesh diagram for cell
luis.ibanez at kitware.com
Fri Mar 18 14:22:14 EST 2005
That sounds like a very interesting project.
Here are several options you may want to try in
order to generate the Voronoi diagram of the cells.
1) If you use a local maximum (or minimum) detection
you can use those points as entry points for the
Danielsson Distance Map, which will compute a
Voronoi partition for you.
For details, please look at the ITK Software Guide
Section 6.6, pdf-page 197, in particular you should
look at the figure 6.30.
2) You could try the Vornoi Segmentation method
That will give you a finer partition than the
one you are looking for. Then you could explore
methods for going back to a coarser partition.
Section 9.4, pdf-page 396.
3) You could directly track the biological cells
using the ITK bio cell classes.
Note that this method will use an overconnected
mesh, so you still will have to simplify the connections
in order to obtain a mesh.
For examples on the use of these classes you may
want to look at:
NOTE that this methods are a bit the opposite of
what you are doing. These methods use the concepts
of cellular behavior in order to perform image
processing. What you are doing is to use image
processing in order to understand cellular behavior.
hash sr. wrote:
> iam working on geometric modeling of the growth process in the embryo
> development. Iam trying study the modeling parameters by studying a video
> showing the cell development in culture. I have generated the frames from
> the video at different times. I would like to generate the voronoi
> representation of the cells. Can anyone tell me what kind of segmentation do
> i have to use? My aim to get the vertices of the voronoi diagram and the
> connectivity from the frame. I would be very glad if someone could help me
> on this.
> Hari Shankar
> Brain Networks Laboratory
> Texas A&M University,College Station
> Insight-users mailing list
> Insight-users at itk.org
More information about the Insight-users