gaps in the pipe modellation

Added by Hillenbrand Matthias almost 3 years ago

Hi Jan,
I have another question on the tree volume detection. I scanned different park trees and in the output I always received a tree model with a lot of gaps (ply-→good) in the pipe modellation so that the volume sometimes is underestimated on a high scale. By the way: are the bad results also integrated in the result table (result from ply-→colour)?
I put two examples in the appedendix. The workflow I used was provided by you and is called <<SimpleTree_Beta_4.33.03_Computree 5.0.054b_plot_fully_automatic_no_deleaving>>
Do you have an idea how to handle this problem?

Also there is one point cloud for that I don’t get any results. In the computing process the steps that are normally taking a lot of time are finished within less than a second (appendix)

Best regards and thanks in advance

Replies (1)

RE: gaps in the pipe modellation - Added by Hackenberg Jan almost 3 years ago

Hi Matthias

I had quickly a look at your clouds. Please try out with the new Version of CT/SimpleTree I will upload the next minutes (4.33.06) the attached script.

a) Your clouds do not have a good quality. A lot of the tree crown is missing. That will cause issues with the allometric scaling approach.
b) In the ply folder named “good” there is only the cylinders which are expected to be good fitted. You will find the other cylinders belonging to the same tree in the folder “bad”. The combined volume of both cylinder sets is the total volume. USe it on your own risk. Better to build your own stem taper/volume prediction function from the incomplete “good” cylinders if you are knowning what to do.