You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First of all, thank you for your excellent work on this project. It's been incredibly valuable to our team.
However, we've encountered an issue during our simulations. When using ALS to perform vertical scans on the constructed forest model, we noticed that almost no ground points are being scanned beneath each tree. We suspect that the mesh for the leaves in our individual tree models might be too large, preventing the scan from reaching the ground effectively.
Could you please provide some suggestions on how to modify the parameters to address this issue? Specifically, we are looking for guidance on which settings we might adjust to reduce the leaf mesh size or improve ground point detection.
Thank you for your assistance!
The text was updated successfully, but these errors were encountered:
Thanks for the kind words, we appreciate hearing that.
In the screenshot, are you using the two trees from data/sceneparts/arbaro? These are created as fairly "low-poly" trees and have unrealistically large leaves (i.e., around 0.9 m in length). This may be the reason for the lack of ground points. We created them with Arbaro, a very similar algorithm is implemented with the "Sapling Tree Gen" add-on in Blender. So you could generate your trees with this software and make sure that the leaves are of realistic size.
As for HELIOS++, you could play around with the full waveform settings, e.g., the beamSampleQuality or the windowSize.
It would of course be helpful to have a real ALS reference point cloud so that you know which proportion of ground returns you are expecting and want to "tune" the simulation to.
Happy New Year!
First of all, thank you for your excellent work on this project. It's been incredibly valuable to our team.
However, we've encountered an issue during our simulations. When using ALS to perform vertical scans on the constructed forest model, we noticed that almost no ground points are being scanned beneath each tree. We suspect that the mesh for the leaves in our individual tree models might be too large, preventing the scan from reaching the ground effectively.
Could you please provide some suggestions on how to modify the parameters to address this issue? Specifically, we are looking for guidance on which settings we might adjust to reduce the leaf mesh size or improve ground point detection.
Thank you for your assistance!
The text was updated successfully, but these errors were encountered: