Jump to content

 
Photo

Bump mapping with LiDAR using ArcGIS

- - - - -

  • Please log in to reply
8 replies to this topic

#1
Matthew Hampton

Matthew Hampton

    Hall of Fame

  • Moderator
  • PipPipPipPipPipPipPip
  • 1,295 posts
  • Gender:Male
  • Location:Portland, Oregon
  • Interests:Playing in the mountains and rivers.
  • United States

We recently published a map service that integrates bump mapping at larger scales. I created the 'bumps' by making a hillshade (LiDARshade?) of the first LiDAR return. Since it's the first surface the laser hits, you get a nice model of treetops, vegetation and buildings. This data is resampled to 3ft pixels for manageability at a regional scale (~20gb).

I added the bare earth surface to illuminate ( :P) the conventional hillshading and a land-cover layer of vegetation for color. If the servers are running slow you can see the buildings draw, but they're generally covered-up by building footprint polygons.

To make the LiDAR shading I increased the azimuth of the light source to ~65º and think it's a rather efficient way to make bumps using ArcGIS.

You can explore other areas by going to http://gis.oregonmetro.gov/metromap and zooming into one of the larger scales. It's my first tiled map service and needs a bit more polish, but I like the direction it's headed.

Attached File  LiDAR_bumpmaps.jpg   148.34KB   170 downloads

co-cartographic creator of boringmaps.com


#2
François Goulet

François Goulet

    Ultimate Contributor

  • Validated Member
  • PipPipPipPipPipPip
  • 688 posts
  • Gender:Male
  • Location:Mille-Isles, Qc
  • Interests:Cartography, History, Graphic Design and almost everything else...
  • Canada

The result is impressive. Well done!



#3
pisgah

pisgah

    Newbie

  • New Member
  • Pip
  • 5 posts
  • United States

We recently published a map service that integrates bump mapping at larger scales. I created the 'bumps' by making a hillshade (LiDARshade?) of the first LiDAR return. Since it's the first surface the laser hits, you get a nice model of treetops, vegetation and buildings. This data is resampled to 3ft pixels for manageability at a regional scale (~20gb).

I added the bare earth surface to illuminate ( :P) the conventional hillshading and a land-cover layer of vegetation for color. If the servers are running slow you can see the buildings draw, but they're generally covered-up by building footprint polygons.

To make the LiDAR shading I increased the azimuth of the light source to ~65º and think it's a rather efficient way to make bumps using ArcGIS.

You can explore other areas by going to http://gis.oregonmetro.gov/metromap and zooming into one of the larger scales. It's my first tiled map service and needs a bit more polish, but I like the direction it's headed.

attachicon.gifLiDAR_bumpmaps.jpg

Great job on the map! Can you give some more detail on creating the 'lidarshade' from first returns? I have first-return points already and am using ArcGIS.


Thanks



#4
Lui

Lui

    Master Contributor

  • Validated Member
  • PipPipPipPip
  • 153 posts
  • Gender:Male
  • Location:Ljubljana
  • Interests:Natural looking maps, hillshading, mountain cartography, LIDAR, TLS, data fusion, photogrammetry, singing, poetry, cycling
  • Slovenia

Ah LIDAR bump -mapping. Always favorite of mine. Your approach is more subtle than my but I like it. Here are some from several years ago:

Clip2.jpg
 
Clip1.jpg
 
Hillshading is a combination of classical one and special one developed just for LIDAR datasets (DSM). 


#5
Lui

Lui

    Master Contributor

  • Validated Member
  • PipPipPipPip
  • 153 posts
  • Gender:Male
  • Location:Ljubljana
  • Interests:Natural looking maps, hillshading, mountain cartography, LIDAR, TLS, data fusion, photogrammetry, singing, poetry, cycling
  • Slovenia

I just take a look at your map and would like to suggest to take out (reclassify) all powerlines from 1st return and to adjust vector data to LIDAR data especially shorelines. Water (clear) doesn't reflect laser beam so water surface created from LIDAR usually has some serious artifacts, especially in shore region and it's not flat.



#6
Matthew Hampton

Matthew Hampton

    Hall of Fame

  • Moderator
  • PipPipPipPipPipPipPip
  • 1,295 posts
  • Gender:Male
  • Location:Portland, Oregon
  • Interests:Playing in the mountains and rivers.
  • United States

I agree about taking out the powerlines from the LAS files  - it's amazing they get captured by the laser beams in the first place!  For the current map service, unclassified 1st return LAS points were converted to a raster - based on their height values.  The result is a high-resolution raster elevation file and I used a traditional "hillshading method" to create the LiDARshading - although I changed the azimuth and altitude from my normal settings to achieve the look I was after.  In addition, I added a bare-earth hillshade to provide the underlying topographic relief shading.

 

Lui, your mention of water and shorelines is quite interesting.  We derived stream-centerlines from the LiDAR data, but for solid bodies of water there are definitely some issues to consider.  I recently attended a forum on the use of green-wavelength LiDAR for topo-bathymetrywerw3.  The test flight used both red and green laser sensors and the results were quite interesting.  You  could combine the data from each sensor to delineate shorelines and since the green wavelength was not reflected, water depth could be captured.

 

There's certainly a bit of research and development needed and lot of variables to consider (water clarity, rapids, etc.), but it's an intriguing new technology.  Attached in an oblique point-cloud image (RGB values mapped to the points) as well as a bathymetric depth gradient from WSI who helped with the flight.

 

Attached File  Screen shot 2013-06-18 at 2.22.24 PM.png   450.09KB   46 downloads 


co-cartographic creator of boringmaps.com


#7
Lui

Lui

    Master Contributor

  • Validated Member
  • PipPipPipPip
  • 153 posts
  • Gender:Male
  • Location:Ljubljana
  • Interests:Natural looking maps, hillshading, mountain cartography, LIDAR, TLS, data fusion, photogrammetry, singing, poetry, cycling
  • Slovenia

I've also had a test project with "water" laser and results in clear water are promising especially in shallow water where multibeam sonar can't work. In not so clear water (ponds, muddy lakes,...) results were not promising. 

I also combine classic bare earth hillshading and custom one. The result is more natural by using some kind of diffusion lightning.



#8
AndrewM

AndrewM

    Contributor

  • Validated Member
  • PipPip
  • 27 posts
  • United Kingdom

Just thought I'd chip in with an article I came across yesterday (via a link on twitter I think) about a new underwater LiDAR system that's being developed in Canada. One of the main applications seems to be to work alongside aerial surveys to enhance/validate their results.

 

I haven't used any data from bathymetric LiDAR myself, but I've worked quite a bit with survey data collected in coastal areas and I know how much of a pain data gaps and reflections due to standing water can be, as well as the survey extent being limited by whatever the low tide level happens to be when they fly it! 



#9
Agnar Renolen

Agnar Renolen

    Contributor

  • Validated Member
  • PipPip
  • 49 posts
  • No Country Selected

Here's my first map produced using LiDAR bump mapping (from 2011):

 

http://www.emap.no/z...011utsn_595.jpg

 

Agnar






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

-->