Accurate 3D comparison of complex topography with terrestrial laser scanner: Application to the Rangitikei canyon (N-Z)

摘要: Abstract Surveying techniques such as terrestrial laser scanner have recently been used to measure surface changes via 3D point cloud (PC) comparison. Two types of approaches pursued: tracking homologous parts the compute a displacement field, and distance calculation between two clouds when cannot be defined. This study deals with second approach, typical natural surfaces altered by erosion, sedimentation or vegetation surveys. Current comparison methods are based on closest require at least one PC meshed severe limitations present roughness elements all scales. To solve these issues, we introduce new algorithm performing direct in 3D. The method has steps: (1) normal estimation orientation scale consistent local roughness; (2) measurement mean change along direction explicit confidence interval. Comparison existing demonstrates higher accuracy our well an easier workflow due absence meshing Digital Elevation Model (DEM) generation. Application rapidly eroding, meandering bedrock river (Rangitikei River canyon) illustrates its ability handle differences complex situations (flat vertical same scene), reduce uncertainty related averaging generate maps levels. We also demonstrate that for high precision survey scanners, total error budget detection is dominated registration roughness. Combined mm-range georeferencing clouds, levels down 6 mm (defined 95% confidence) can routinely attained situ over ranges 50 m. provide evidence self-affine behaviour different surfaces. show how this impacts vectors scaling level detection. implemented freely available open source software package. It operates cases simpler more robust alternative DEM differencing 2D cases.