analyse local thickness of laserscan-data || road surface

[toxicmag]

Dear CC-users,

i am searching for an elegant way of a "local pointcloud analysis".
I attached some example screenshots and looking forward to read your suggestions.

The scan is from an airborne lidar and i'd like to assess the quality of the scan on a road surface.

A paved road should be a sharp / crisp surface in the scan data.
The more noise and thickness the data shows the less useful it is.

What in your eyes could be a good approach to easily see areas of higher thickness, its thickness values and so on...?
From my point of view a good approach would be a segmentation first (subdivide into a grid or define a radius) and then a local analysis for each segment.

What would be your suggested steps for it?
;)

Thanks and greetings

Alex

[daniel]

You could try to compute the local "roughness" (https://www.cloudcompare.org/doc/wiki/i ... c_features)

Or use the Rasterize tool, and export the grid as a cloud with 'height std. dev.' and 'height range' in each cell:

rasterize.JPG (20.05 KiB) Viewed 1444 times