Centre de documentation
scientifique

Streambank topography: an accuracy assessment of UAV-based and traditional 3D reconstructions / Benjamin U. Meinen in International Journal of Remote Sensing IJRS, vol 41 n° 1 (01 - 08 janvier 2020)  

Public [article]  inInternational Journal of Remote Sensing IJRS > vol 41 n° 1 (01 - 08 janvier 2020) . - pp 1 - 18 Titre : Streambank topography: an accuracy assessment of UAV-based and traditional 3D reconstructions Type de document : Article/Communication Auteurs : Benjamin U. Meinen, Auteur ; Derek T. Robinson, Auteur Année de publication : 2020 Article en page(s) : pp 1 - 18 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques [Termes IGN] carte topographique [Termes IGN] cours d'eau [Termes IGN] détection de changement [Termes IGN] érosion [Termes IGN] hétérogénéité spatiale [Termes IGN] image à très haute résolution [Termes IGN] image captée par drone [Termes IGN] modèle numérique de surface [Termes IGN] photogrammétrie aérienne [Termes IGN] reconstruction 3D [Termes IGN] rive [Termes IGN] structure-from-motion [Termes IGN] télémètre laser terrestre Résumé : (auteur) Highly accurate digital surface models are an essential part of change-over-time analyses for monitoring erosion processes. Streambank topography presents a unique challenge for surface mapping due to dense riparian vegetation, canopy cover, and rapidly changing elevation values. The spatial heterogeneity of stream corridors has made the calculation of streambank erosion across larger spatial extents difficult. Contemporary technologies such as terrestrial laser scanners (TLS) and unmanned aerial vehicles (UAVs) offer new approaches for streambank topography mapping at very high spatial resolutions across varying spatial extents. To evaluate the accuracy of different technologies for streambank topography mapping, we compared streambank surface models derived via a UAV using structure-from-motion and from traditional aerial photogrammetry (i.e. Southwestern Ontario Orthoimagery Project; SWOOP) to that of a TLS benchmark across seven streambank segments. Additional comparisons were made for 22 manually measured stream transects to that of a TLS benchmark. Compared to our benchmark, the UAV-derived streambank surface model was the most accurate with an average root-mean-square-error of 0.104 m. Errors in the UAV surface model were correlated with georeferencing error. The UAV had an average 52% success rate for reconstructing the streambank topography across all field campaigns and was able to map up to 2037 m of streambank in one hour. The streambank surface model derived from traditional aerial photogrammetry and manual transect measurements had average root-mean-square-errors of 0.238 m and 0.274 m respectively. Both aerially-derived surface models tended to over measure elevation values compared to the TLS, whereas manual transect measurements consistently under measured elevation. Numéro de notice : A2020-208 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1080/01431161.2019.1597294 Date de publication en ligne : 26/03/2019 En ligne : https://doi.org/10.1080/01431161.2019.1597294 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94890 [article]