Extract meshes by texture color / Trees on Lidar surface

Hello Forum,

I have the following problem:

I have some relatively large topographical meshes from LIDAR Data with high res orthographic photos mapped them. Now I want to populate ares with vegetation, which can be identified on the orthophotos (…the green stuff) with real trees. I already made some successful test with the awesome Grow plugin by Cam Newnham. Workflow is to manually cut out mesh areas with trees (AKA forests…) and use these as a base for block distribution.
While this works for some of the areas, it is not applicable for others which are to small and scattered, it would take me ages to do this by hand…
I wonder if someone might have an Idea how to approach this problem, since we have the color information from the photo (everything wich is green…) this could be transformed e.g. to a black and white map in PS to define areas for placing trees / not placing trees… maybe OCR would work…?
Thanks for any ideas / feedback

best

Andreas

a picture of the textured mesh:

a picture of experimental tree distribution:

Hi Andreas - I imagine it should be possible to extract the mapped color information at each mesh vertex and extract the adjacent faces if within some range… I don’t know… emphasis on the ‘imagine’, I have not looked into it yet, but it might be an interesting puzzle to solve in a script.

-Pascal

1 Like