Lastools workflow advices - Intensive greening potential evaluation based on LIDAR Data (.las)

[Maeva]

Dear Lastools Community!

I am a beginner in GIS-softwares and LAStools and I would like to ask you recommendations for my research workflow with QGIS/LASTOOLS.

The goal of my study is to evaluate the potential of intensive greening for the city of Rio de Janeiro (to find out which roofs have an inclination which is less than 5 degrees) .

I have received the :

- LIDAR: las. files of the city of Rio

- Building footprints (with elevation but no pitch)

Here my workflow so far:

1) I have started to work on Lastools following the tutorial"tiles and prepare" : https://rapidlasso.com/2013/10/13/tutorial-lidar-preparation/.

2) The output (tiles classified) of this tutorial is then transformed in shp files with las2shp (with the LASTools - GUI version) - I keep the elevation, the intensity and classification of the points as attributes.

3) In QGIS, I clip the point shp files with the building footprints. I have obtained only the points which fall into the polygons of my roof.

So here my questions:

a) Does this preparation workflow sound satisfying for you? Should I add other steps to be more precise?

b) What would you recommend as next steps with LASTools to find out the most precisely the angle of each roofs with these points and the roof polygons?

I am convinced there are many way/methodology to find the roof inclination and I would like to know the most precise one following your precious advices :)

PS: thank you Martin Isenburg for the great work, I enjoy your tool very much!

Kind regards from Rio

xxx

[Steeve Brissette]

I would personally try to use lasclip, and isolate (clip) every building in separate files. Than You could run over each tile (loop) and look for the lowest and highest point which you than use to calculate the pitch. You might want to buffer down the polygons of your building footprint by half a metre of so, to make sure you don’t end up with the base of the building, or some vegetation, which would definitely screwup your pitch.

Sorry, I am actually walking on the street, so I don’t have any code snippet for you.

That would be my first approach..

Steeve Brissette 

Civil/Geomatics Technician 

On May 20, 2018, at 7:17 AM, Maeva <dang....@gmail.com> wrote:

Dear Lastools Community

--
Download LAStools at
http://lastools.org
http://rapidlasso.com
Be social with LAStools at
http://facebook.com/LAStools
http://twitter.com/LAStools
http://linkedin.com/groups/LAStools-4408378
Manage your settings at
http://groups.google.com/group/lastools/subscribe

[Peter Guth]

Hi,

I would create a DSM grid from the point cloud, make a slope map from the DSM, mask the slope maps to just have the buildings, and then look at the slopes for each building.

--

Peter L. Guth
Professor, Dept Oceanography, USNA

[Steeve Brissette]

Hi...

... how many buildings you have in the city of Rio.. your aim is probably to automate the workflow, I would suggest to look into lasboundary as well, you might find that tool useful if you want to create a vector representation of the roofs after you clipped them, which will also contain Z information (real value). 

Now you need to know which slope belongs to each roof, you probably have an identifier in your footprint file that you can use to name your buildings (clip)... then I would generate a report using that name, and the pitch, all in a text file. Using an automated workflow, and Lastools.

You might need other libraries though, like OGR/GDAL.

Good luck.

Steeve Brissette 

Civil/Geomatics Technician

[Maeva]

Dear Peter & Steeve,

thank you for your precious advices!

1) I have started to work with the Lidar data and tried to generate by myself the roof surface polygons (using Tutorial 1 and 2 of martin even if I already have some building footprints provided by the city).

The results I have obtained is not very good (some vegetation is classified as buildings and the roofs units are merged together). 

I think I should adapt a few parameters. I have played with the concavity and the search area size (2 and 3) but I didn't get better results. Do you please have some hints to produce more accurate polygons?

2) @ Peter Guth:

Thank you for your help!

I have a lot of buildings for this district of Rio. How would you automate the workflow using DSM & Slope Map? What tool would you use to identify fast the slope for each roof without checking them one by one?

3) @ Steeve:

Thank you for your answers!

I will try this workflow as well and keep you informed :)

[Steeve Brissette]

Is this possible to have a sample of your data? LAZ and SHP?

Steeve Brissette