Questions and algorithm
tool for manage and process TLS lidar data from forest environment
Brought to you by:
janekt,
m-a-r-t-in
Hello,
my name is Felix Behrendt and I´m a student (Geoninformatics) at the Friedrich Schiller University, Jena, Germany. Currently I am working on my Master Thesis about comparing point clouds of a tree. Part of the work is the reconstruction of 3D Models. Your software provides an algorithm. Here I want to thank you, because your model works for my data, but I have a few questions about the algorithm and the software.
In the Wiki (https://github.com/janekT/3DForest/wiki/10.-QSM-models-and-tree-sortiments) and the paper (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0176871) is roughly explained. Is there a more accurate documentation about the algorithm? Or could you explain this sentence:
"If all neighbors are connected into one piece, then its is part of the same stem."
(How is a connection defined?)
Another point is about the in the wiki mentioned input distance. I can´t set this value. How is it calculated ?
My last question is about the software. The software plots the 3D model. Is it possible to extract this model somehow? (Currently I can only export the extracted parameter)
Many thanks for your help and your provided software.
With best regards,
Felix Behrendt
Hi Felix,
Thanks for using our software.
Explanation of stem connection is hard to explain in words, but I try the
best. If you imagine point for which you are looking for neighbors (lets
name them as N), now you can test if they are connected together of if they
are creating two separate parts. You can on attached figure. for point
(red) you find points in its neighborhood (point in blue circle), now
starts analyzing those point is they are all connected together or
not.(green circle). On figure you can see that it seem that we have two
separate parts - left and right points that are not connected - possible
branch.
Input Distance = input minimal coverset.
For now it possibly export only parameters and not models. But I believe we
will provide such functionality in next release.
Jan
po 18. 11. 2019 v 15:06 odesílatel noreply@sourceforge.net napsal: