Gridding lidar data - artifacts and seam lines

geojules

Hello,

I am using Global Mapper (+LIDAR module) to grid a large amount of lidar data. The AOI is 2,361 km2 and I have 1838 LPC tiles at a point density of about 7.24 pts/m2. Needless to say, with this amount of data, I need to use a tiling scheme when I run the 'Create Elevation Grid from 3D Vector Data' tool. I think this is where my issues originate.

To begin with I used the Binning method to grid the lidar data. This method reliably produces 'striping'/'corn-row' artifacts. These artifacts only appear in the top row or left most column of the tiling scheme, depending on the arrangement of tiles. For example, the stripes align north-south (screen1.png) in the top tile of a 2x1 (rows x columns) tiling scheme or east-west (screen2.png) in left tile of a 1x2 tiling scheme. This pattern is still present in tiling schemes with more tiles, e.g. 4x5.

Because the location/direction of these artifacts change with the same data, but a different tiling scheme, I am certain they are not inherent to the data (which undergoes rigorous QA/QC).

I found that gridding the data with the TIN method does NOT produce these artifacts. This is not an ideal workaround, however, because of the increased processing time and the creation of visible 'seamlines' at the boundary between tiles. I've read elsewhere on this forum that adding the resulting DEM files (.gmg) to a map catalog will remove these seamlines (see here). I find this not to be true. Even when I add the tiles to a map catalog the seamlines are visible (screen3.png). Furthermore, when I export these files to GeoTIFF and bring into ArcGIS, these artifacts are visible.

Please let me know if I am missing something. I would really like to use the Global Mapper Lidar module for this workflow, but these artifact issues in the resulting grids are prohibitive of that.

Thanks,
Julian

bmg_bob

Hello,

I suggest that you contact Blue Marble Support directly via email (geohelp@bluemarblegeo.com) to address this issue. Please include a detailed description of your work flow, including screen captures of any options screens and a small dataset that can be used to reproduce the issue. Thanks.

Cheers,
Bob

geojules

Thank you Bob, I will send this issue along to geohelp@bluemarblegeo.com

VincentR

Hi @geojules

I've experienced similar artifacts (striping while binning and not when using triangulation), have reported this to Blue Marble Support by mail, they have opened case 17616.

This has not been resolved yet, so please let me know if you find anything...

Kind regards,

Vincent

DerrickW

I've been having issues for a couple months with creating point clouds from DSMs (in geotiff format).  Resulting point cloud ends up being in strips with nodata between. Related?