Wood/Leaf Classification #29
-
Is rayextract (or any other function) able to separate out the woody parts of the tree from the leafy parts of the tree? If not, are there any tools that others have used for wood/leaf classification on plot-level forest scans? Thanks for the help! |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 2 replies
-
Hi Antra, The output cloud's red channel is proportional to lidar return intensity (see ray import --max_intensity to scale this) and works if your input cloud has a well-calibrated return intensity. It has only been tested on the Velodyne lidar, so may require some tweaking to get right for other lidars. The output cloud's green channel is proportional to the cylindricality of each neighbourhood of points. You can then split the coloured cloud based on a linear combination of red and green using raysplit xxx_coloured.ply colour 0.2,0.1,0 for instance. That last vector is the plane vector in rgb space, so for instance 0,0.1,0 would split around just green=0.1. I don't know of other available separation methods but I'm sure there are some around. |
Beta Was this translation helpful? Give feedback.
-
I think you're after a method like FSCT. |
Beta Was this translation helpful? Give feedback.
Hi Antra,
there is a basic method in raycloudtools: raycolour xxx.ply branches
The output cloud's red channel is proportional to lidar return intensity (see ray import --max_intensity to scale this) and works if your input cloud has a well-calibrated return intensity. It has only been tested on the Velodyne lidar, so may require some tweaking to get right for other lidars.
The output cloud's green channel is proportional to the cylindricality of each neighbourhood of points.
You can then split the coloured cloud based on a linear combination of red and green using raysplit xxx_coloured.ply colour 0.2,0.1,0 for instance. That last vector is the plane vector in rgb space, so for instance 0,…