Digital Terrain Models Generated with Low-Cost UAV Photogrammetry: Methodology and Accuracy

Abstract

Digital terrain model (DTM) generation is essential to recreating terrain morphology once the external elements are removed. Traditional survey methods are still used to collect accurate geographic data on the land surface. Given the emergence of unmanned aerial vehicles (UAVs) equipped with low-cost digital cameras and better photogrammetric methods for digital mapping, efficient approaches are necessary to allow rapid land surveys with high accuracy. This paper provides a review, complemented with the authors’ experience, regarding the UAV photogrammetric process and field survey parameters for DTM generation using popular commercial photogrammetric software to process images obtained with fixed-wing or multicopter UAVs. We analyzed the quality and accuracy of the DTMs based on four categories: (i) the UAV system (UAV platforms and camera); (ii) flight planning and image acquisition (flight altitude, image overlap, UAV speed, orientation of the flight line, camera configuration, and georeferencing); (iii) photogrammetric DTM generation (software, image alignment, dense point cloud generation, and ground filtering); (iv) geomorphology and land use/cover. For flat terrain, UAV photogrammetry provided a horizontal root mean square error (RMSE) between 1 to 3 × the ground sample distance (GSD) and a vertical RMSE between 1 to 4.5 × GSD, and, for complex topography, a horizontal RMSE between 1 to 7 × GSD and a vertical RMSE between 1.5 to 5 × GSD. Finally, we stress that UAV photogrammetry can provide DTMs with high accuracy when the photogrammetric process variables are optimized.

Conventional and UAV-Based Aerial Surveys for Long-Term Monitoring (1954–2020) of a Highly Active Rock Glacier in Austria

“…The general workflow consisted of 1) data import, 2) bundle block adjustment (providing a sparse point cloud), 3) computation of the dense point cloud applying a multiple-view stereo algorithm, 4) meshing (2D), 5) DOP generation, 6) DEM generation, 7) export of results, and 8) reporting (cf. Westoby et al, 2012;Jiménez-Jiménez et al, 2021). Workflows for registration/georegistration of multi-temporal UAV-based image data avoiding GCPs and supporting surface change detection have already been proposed by different authors (Dall'Asta et al, 2017;Feurer and Vinatier, 2018;Chudley et al, 2019;Cook and Dietze, 2019;de Haas et al, 2021;Śledź et al, 2021;Vivero et al, 2021).…”

“…The significance level of the volumetric change, i.e., of the mean surface height change in the catchment area, was not set by the standard error of the mean but by the standard deviation of the residual change (cf. Jokinen and Geist, 2010;Jiménez-Jiménez et al, 2021). The first approach would have given a too optimistic result.…”

GIS-Based Landslide Susceptibility Mapping of the Circum-Baikal Railway in Russia Using UAV Data

“…As soil commonly has reflectance in the red spectrum, it leads to "soil noise pixel"; using orange light allows us to avoid the blurring effect and clearly discern boundaries of soil and vegetation. It is important to precisely highlight vegetation cover as its presence may reduce vertical accuracy and cause distortions of the DTM [29]. that the high-resolution imagery of the sub-vertical sections of the slopes makes it possible to significantly detail such parameters as terrain roughness, in comparison with the nadir imagery-both UAV and satellite.…”

3D Dense Reconstruction of Road Surface from UAV Images and Comparison of SfM Based Software Performance

“…The coordinate information of the images provides spatial accuracy to have a certain level of accuracy in 3D digital models with the UAV measurement technique (Jiménez-Jiménez et al 2021). This may require the collection of ground control points (GCPs) commonly used in practice.…”

“…Therefore, several methods and technologies have been introduced to obtain road surface and its environment. One common way is to extract road surfaces from three-dimensional (3D) data (Biçici and Zeybek, 2021;Jiménez-Jiménez et al 2021;Zeybek and Şanlıoğlu, 2019). There are several technologies to collect 3D data.…”