I have a 3D image with different labelled regions, I would like to extract the voxels defining the surface between regions. I would like to have one voxel wide surface.
I found the chain code algorithm for boundary tracing in 2D images, is there a similar algorithm in 3D?
Thank you.
Dear Amal Amami, Kurtosis is one of the classical estimators of HOS. It is a measure of heaviness of the tails in a distribution. Negative kurtosis, which indicates less peaked and wide distribution with short tails, is referred as ‘platykurtic’ and positive kurtosis that indicates more peaked distributions with fat tails is referred as ‘leptokurtic’. Normal distributions with zero kurtosis are referred as ‘mesokurtic’. So you can prefer this !!!
Dear Thomas ,
That is what i have done but i am getting more than one voxel on the boundary. I would like to obtain a one voxel wide contours.
Hi Amal,
you can use the code of http://www.ipol.im/pub/art/2014/74/?utm_source=doi which will output the set of surfels defining the boundary according the chosen topology definition.
Then you deduce the inside voxel as standard coordinates.
More directly you can use the 3dVolBoundaryViewer tool of the DGtalTools project from the DGtal library and insert the export of the voxel which are displayed.
Best
http://www.ipol.im/pub/art/2014/74/?utm_source=doi
https://github.com/DGtal-team/DGtalTools
https://github.com/DGtal-team/DGtal
- Badges
- Science topic
- Similar topics
- Mathematics