Centre de documentation
scientifique

Spatial accuracy of UAV- derived orthoimagery and topography: Comparing photogrammetric models processed with direct geo-referencing and ground control points / Chris H. Hugenholtz in Geomatica, vol 70 n° 1 (March 2016)  

Public [article]  inGeomatica > vol 70 n° 1 (March 2016) . - pp 21 - 30 Titre : Spatial accuracy of UAV- derived orthoimagery and topography: Comparing photogrammetric models processed with direct geo-referencing and ground control points Type de document : Article/Communication Auteurs : Chris H. Hugenholtz, Auteur ; Owen Brown, Auteur ; Jordan Walker, Auteur ; et al., Auteur Année de publication : 2016 Article en page(s) : pp 21 - 30 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications photogrammétriques [Termes IGN] acquisition de données [Termes IGN] analyse comparative [Termes IGN] capteur aérien [Termes IGN] drone [Termes IGN] géoréférencement direct [Termes IGN] point d'appui [Termes IGN] positionnement cinématique en temps réel [Termes IGN] positionnement inertiel [Termes IGN] positionnement par GNSS [Termes IGN] positionnement par GPS [Termes IGN] qualité des données [Termes IGN] réalité de terrain [Termes IGN] topographie Résumé : (auteur) Mapping with unmanned aerial vehicles (UAVs) typically involves the deployment of ground control points (GCPs) to georeference the images and topographic model. An alternative approach is direct georeferencing, whereby the onboard Global Navigation Satellite System (GNSS) and inertial measurement unit are used without GCPs to locate and orient the data. This study compares the spatial accuracy of these approaches using two nearly identical UAVs. The onboard GNSS is the one difference between them, as one vehicle uses a survey-grade GNSS/RTK receiver (RTK UAV), while the other uses a lower-grade GPS receiver (non-RTK UAV). Field testing was performed at a gravel pit, with all ground measurements and aerial surveying completed on the same day. Three sets of orthoimages and DSMs were produced for comparing spatial accuracies: two sets were created by direct georeferencing images from the RTK UAV and non-RTK UAV and one set was created by using GCPs during the external orientation of the non-RTK UAV images. Spatial accuracy was determined from the horizontal (X,Y) and vertical (Z) residuals and root-mean-square-errors (RMSE) relative to 17 horizontal and 180 vertical check points measured with a GNSS/RTK base station and rover. For the two direct georeferencing datasets, the horizontal and vertical accuracy improved substantially with the survey-grade GNSS/RTK receiver onboard the RTK UAV, effectively reducing the RMSE values in X, Y and Z by 1 to 2 orders of magnitude compared to the lower grade GPS receiver onboard the non-RTK UAV. Importantly, the horizontal accuracy of the RTK UAV data processed via direct georeferencing was equivalent to the horizontal accuracy of the non-RTK UAV data processed with GCPs, but the vertical error of the DSM from the RTK UAV data was 2 to 3 times greater than the DSM from the non-RTK data with GCPs. Overall, results suggest that direct georeferencing with the RTK UAV can achieve horizontal accuracy comparable to that obtained with a network of GCPs, but for topographic measurements requiring the highest achievable accuracy, researchers and practitioners should use GCPs. Numéro de notice : A2016-433 Affiliation des auteurs : non IGN Thématique : IMAGERIE/POSITIONNEMENT Nature : Article DOI : 10.5623/cig2016-102 En ligne : https://doi.org/10.5623/cig2016-102 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=81330 [article]