Using Terrestrial Laser Scanning Data to Validate 3DEP

Submitted by atripp on Fri, 04/03/2020 - 08:42

To validate the quality of light detection and ranging (lidar) data collected for the 3D Elevation Program (3DEP), the USGS uses methods detailed in “Accuracy Standards for Digital Geospatial Data” developed by the American Society for Photogrammetry and Remote Sensing (ASPRS). These methods involve using survey-grade global positioning system (GPS) checkpoints for validation on the bare earth surface spread throughout the project site in flat open areas and under vegetation. While this method is sufficient for evaluating the bare earth digital elevation models (DEMs), it is inadequate to understand the quality of the lidar point cloud data for points above the ground, such as for vegetation and buildings. As a result, the end user may incorrectly assume that the root mean square error (RMSE) derived for the bare earth DEM applies to every lidar point in the dataset. This pilot project used survey-grade terrestrial laser scanning (TLS) data to compare and assess all lidar points collected for 3DEP in plots spread throughout several projects. This project will compare the point clouds, statistical measures for goodness of fit, and other metrics to help guide future ground truthing. This method could help better inform and explain the uncertainties in lidar point cloud data collected for future 3DEP surveys.

Example graphic showing Z differences between TLS and 3DEP data.

Author Name
Jason Stoker
Author Email