DTM generation from dense matching photogrammetric point cloud for hilly and vegetated terrain

180 views
Skip to first unread message

tabish....@gmail.com

unread,
Sep 6, 2017, 9:53:58 PM9/6/17
to LAStools - efficient tools for LiDAR processing
Hello Martin,

I'm currently working on a rail alignment project and require accurate cross sections for the design of slope stability solutions along the rail corridor.

As operations are in a remote and hostile terrain we experimented by deploying a UAV equipped with 20 Megapixel 1 inch CMOS sensor to generate an output with 40 million data points for an area of roughly 1.2 sqkm with GSD of 4.5 cm.

 We are stuck on the ground classification stage wherein areas of dense vegetation are been left as holes in the data set and hence our contour plots and cross sections are been left open.

We have been trying to classify ground using a trial version for auto classification tools available in the market.

But the results aren't satisfactory. So can you guide me on how the above goal can be achieved in Rapidlasso?


Best,
Tabish Kalsekar
Mumbai, India

 




Tobias K Kohoutek

unread,
Sep 7, 2017, 8:58:05 AM9/7/17
to LAStools - efficient tools for LiDAR processing
Dear Tabish,

you can by a camera with 100 Megapixel and still will be unable to see what's inside shaded areas. A camera isn't an active sensor.

So without any additional informatio like measurements in the field it's impossible what you want to achieve.

A guide how to you could have found easily, just by checking rapidlasso.com webpage. Something like:

Cheers,
Tobias

Martin Isenburg

unread,
Oct 16, 2017, 4:36:39 AM10/16/17
to LAStools - efficient command line tools for LIDAR processing
Hello Tabish,

as Tobias pointed out, we have a number of articles on how to use LAStools to get reasonable DTMs from dense-matching photogrammetry on our company blog that often provide the data and guide you through step by step:


Given the density of these point clouds I usually advice to run the entire ground finding pipeline only on a thinner subset of points and then desnify it later by adding all the points that are within +/- 2 or +/- 5 cm of this thinner ground TIN to the final set of ground points.

It seems you need external ground points for a pipeline similar to the second of the blog articles listed above.

Regards.

Martin @rapidlasso




Reply all
Reply to author
Forward
0 new messages