- An integrated sUAS and SfM approach is used to generate topographic data.
- The SfM dataset is compared with a total station ground survey.
- Causes of vertical difference between the two datasets are investigated.
- The integration of a combined sUAS and SfM approach is discussed.
Abstract: Novel topographic survey methods that integrate both structure-from-motion (SfM) photogrammetry and small unmanned aircraft systems (sUAS) are a rapidly evolving investigative technique. Due to the diverse range of survey configurations available and the infancy of these new methods, further research is required. Here, the accuracy, precision and potential applications of this approach are investigated. A total of 543 images of the Cwm Idwal moraine–mound complex were captured from a light (< 5 kg) semi-autonomous multi-rotor unmanned aircraft system using a consumer-grade 18 MP compact digital camera. The images were used to produce a DSM (digital surface model) of the moraines. The DSM is in good agreement with 7761 total station survey points providing a total vertical RMSE value of 0.517 m and vertical RMSE values as low as 0.200 m for less densely vegetated areas of the DSM. High-precision topographic data can be acquired rapidly using this technique with the resulting DSMs and orthorectified aerial imagery at sub-decimetre resolutions. Positional errors on the total station dataset, vegetation and steep terrain are identified as the causes of vertical disagreement. Whilst this aerial survey approach is advocated for use in a range of geomorphological settings, care must be taken to ensure that adequate ground control is applied to give a high degree of accuracy.
Tonkin, T.N., Midgley, N.G., Graham, D.G. and Labadz, J.C. (2014). The potential of small unmanned aircraft systems and structure-from-motion for topographic surveys: A test of emerging integrated approaches at Cwm Idwal, North Wales. Geomorphology, 226, pp. 35-43.
Publisher link: http://www.sciencedirect.com/science/article/pii/S0169555X14003857 (subscription required)
Preprint: https://dspace.lboro.ac.uk/2134/19115 (no subscription required)