How accurate a topography can be reconstructed using unmanned aircraft system and structure from motion?

Authors: Yingkui Li*, University of Tennessee
Topics: Geomorphology, Remote Sensing, Field Methods
Keywords: unmanned aircraft systems, structure from motion, terrestrial laser scanner, topography reconstruction
Session Type: Poster
Day: 4/5/2019
Start / End Time: 8:00 AM / 9:40 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded


Recent development in unmanned aircraft systems (UAS) and structure from motion (SFM) provides a rapid, efficient, and low-cost means to map detailed topography and quantify earth surface processes in a much finer (up to cm-level) resolution. The work presented in this poster assesses the accuracy of using a low-cost UAS to map a hillslope topography based on the comparison with the topography generated using a terrestrial laser scanner. We surveyed a rill-dominated hillslope in Loudon County, Tennessee using a DJI Phantom 3 Advance drone and a FARO Focus 3D X330 terrestrial laser scanning (TLS). The drone was flown about 15 m above the ground and collected photos with four different flight missions at the same flight height: one nadir and three oblique missions with an angle of 30 degree from three perspectives. The 3D point clouds were generated using the photos collected by different combinations of these flight missions. An optimized workflow was developed to register different point clouds to the sub-centimeter level and facilitate comparisons in CloudCompare and ArcGIS. The results indicate that the UAS-generated point clouds can reach up to < 5 mm difference from the TLS-produced point clouds and the inclusion of oblique photos can significantly improve the topography reconstruction in steep slope areas.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login