1

This will be a little long-winded but I'm including all information I consider relevant.

Problem:

I have received a data set from a municipality that is a 3-layer LiDAR JP2 DEM and need to extract ground contours at 1 m intervals for a specified area (more below).

Programs available:

  • QGIS 2.18 Las Palmas with LASTools add-on installed.
  • GRASSgis 7.4.4 with GDAL.
  • Civil3D 2017.
  • Land Desktop 2004.

Additional Information:

I normally work with LiDAR in .laz/.las formats and use LASTools through QGIS to extract the ground points with laszip and then create a .tin for use in Civil3D through Recap360 (.rcap import).

I can load the .JP2 DEM into GRASS with a proper spacial reference (UTM20), but it comes in as a 3-layer raster map with color naming being the only difference (R/G/B)

When I run r.contour to create a vector layer capable of being exported to .dxf, the returned vector polys are very noisy and unusable. I've tried a cheap work-around by increasing the minimum points required per contour but it's still unusable. I've also tried exporting the DEM .jp2 into a gtiff file but it's unrecognized by Civil3D (Civil3D takes .dem, .tif, .asc, .txt, and .adf for surface creation, but seemingly won't take a gtiff).

Assumed solutions:

  • I need to first merge the 3 band of scans, then extract the ground contours through vector .dxf, or
  • Combine the 3 bands and export as a singular .dem to be directly imported into Civil3D, or
  • Combine the 3 bands and export as a .shp, import .shp into LandDesktop, and then use the auto-created .tin file directly in Civil3D.

Any help is greatly appreciated. I can also upload the .zip of the scan to a cloud storage if someone wants to try their hand directly (it's public non-proprietary LiDAR set), but would like to know the solution regardless as I will likely need to do this again in the near future.

0

I can load the .JP2 DEM into GRASS ... , but it comes in as a 3-layer raster map with color naming being the only difference (R/G/B)

You do not have a DEM (elevation data), but RGB bands. This means it is impossible to extract contours from that image. The 'unusable contours' are consequence from interpolating color data (instead of elevation).

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.