small scale cloud points by SfM - ground classification

67 views
Skip to first unread message

Marco Palma

unread,
Nov 22, 2016, 5:58:19 AM11/22/16
to LAStools - efficient tools for LiDAR processing
Hi,
  I am testing the lastools with cloud points generated by underwater SfM surveys. 
Due to the high complexity of the ground morphology, the gorgonians "forest" that is growing on it and the very small dimension of the study area, if compared with the normal usage, I am facing several problems to get a good identification of the ground using lasground. 

The overall area is 10*3 meters; my "trees" reach the maximum height of 80cm and the floor is not flat but it looks like an high steep mountain cliff.

Do you have any suggestion for a better calibration of the tool?
Cheers
Marco

Martin Isenburg

unread,
Nov 24, 2016, 8:36:04 AM11/24/16
to LAStools - efficient command line tools for LIDAR processing
Hello,

you will have to miniaturize the parameters for lasground as well. Simply try picking parameters that are correspondingly shrunk (assuming you have a very high return density given such a small area).


Maybe try a '-step 0.5' and a '-fine' preprocess with '-bulge 0.1' and '-offset 0.01' but without you sharing the data there is little hope that someone can volunteer a good set of parameters.

Sometimes lasthin (lowest) or lasgrid (lowest) or lascanopy (p01 or p05) also provide viable ways to find a ground surface but it really depends on your data.

Regards,

Martin @rapidlasso

Reply all
Reply to author
Forward
0 new messages