search

tiagodc / TreeLS

R functions for processing individual tree TLS point clouds
GNU General Public License v3.0
82 stars 27 forks source link

Add or remove misclassified result with on RGL click #46

Open mzeybek583 opened 3 years ago

mzeybek583 commented 3 years ago

Is there any possibility or package to improve automated inventory extraction after automation and detection missed tree on RGL mouse click? Due to challenges in the various environment, misclassified commissioned or omissioned tree detection and inventory parameters corrections (add/remove) would be very useful. Any idea about it? Thanks. Mustafa

spokswinski commented 3 years ago

The map.pick function within treeMap allows you to add missed trees. It does not allow you to declassify. Two methods to prevent misclassification of non-trees include adjusting the parameters on the algorithm you choose to be more strict then add missed trees manually using the map.pick function OR run fastPointMetrics and filter_poi with high Verticality (~80-95) and low Eigentropy (~<0.3) then using the filtered point cloud to find probable tree regions using treeMap, then use the resulting map against the original pointcloud to classify tree regions using treePoints, THEN stemPoints.