Accurate automated assessment of gully cross‐section geometry using the photogrammetric interface FreeXSapp

• Published in: Earth surface processes and landforms 2018-06, Vol.43 (8), p.1726-1736
• Main Authors: Castillo, C., Marín‐Moreno, V.J., Pérez, R., Muñoz‐Salinas, R., Taguas, E.V.
• Format: Article
• Language: English
• Subjects: Automation; Depth perception; Dimensions; Erosion; Gully erosion; Lidar; Methods; Operating systems; Orientation; Photogrammetry; Scaling; Smartphones; Software
• ISSN: 0197-9337; 1096-9837
• DOI: 10.1002/esp.4341

During recent years, 3‐D techniques such as LiDAR and Structure‐from‐Motion (SfM) photogrammetry have been increasingly used for gully erosion assessment. However, innovative image‐based approaches based on these advances may also be used to provide accurate cross‐sectional measurements which are less expensive and time‐demanding. In this work, we present the FreeXSapp methodology, a new piece of freely‐available software based on existing SfM tools (MicMac and PMVS2) and augmented‐reality targets (ArUco markers) which performs the automated 3‐D reconstruction, scaling, orientation and analysis of gully cross‐sections (XSs) from images taken from the gully margin using a smartphone camera. As a field application, the volume of a 60‐m‐long medium‐size gully was evaluated, where a total of 10 XSs were measured and analyzed in approximately 30 min. The relative accuracy in estimating width and depth dimensions was in the order of 0.5%, with a precision ratio (relative to the camera–XS distance) of ~1500. Overall, using this methodology showed excellent performance in terms of time and cost requirements when compared with typical 3‐D and conventional 2‐D techniques. The FreeXSapp interface is downloadable for free for Windows operating systems at http://www.uco.es/users/ccastillo/freexsapp. Copyright © 2018 John Wiley & Sons, Ltd. We present the FreeXSapp methodology, a new piece of freely‐available software based on existing SfM tools (MicMac and PMVS2) and augmented‐reality targets (ArUco markers) which performs the automated 3‐D reconstruction, scaling, orientation and analysis of gully cross‐sections (XSs) from images taken from the gully margin. This methodology showed excellent performance in terms of accuracy, time and cost requirements when compared with typical 3‐D and 2‐D techniques. The FreeXSapp interface is downloadable for free for Windows OSs at http://www.uco.es/users/ccastillo/freexsapp.