Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland
Real-Time/Post-Processing Kinematic (RTK/PPK) technology has been widely applied in Unmanned Aerial Vehicle (UAV) photogrammetry in glaciological research. Considering that ground control points (GCPs) cannot be set on glaciers, evaluating the impacts of one-sided distribution is essential. In this...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/21/3934 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850062400155287552 |
|---|---|
| author | Haiyan Zhao Gang Li Zhuoqi Chen Shuhang Zhang Baogang Zhang Xiao Cheng |
| author_facet | Haiyan Zhao Gang Li Zhuoqi Chen Shuhang Zhang Baogang Zhang Xiao Cheng |
| author_sort | Haiyan Zhao |
| collection | DOAJ |
| description | Real-Time/Post-Processing Kinematic (RTK/PPK) technology has been widely applied in Unmanned Aerial Vehicle (UAV) photogrammetry in glaciological research. Considering that ground control points (GCPs) cannot be set on glaciers, evaluating the impacts of one-sided distribution is essential. In this study, 8571 images were captured at Sermeq Avannarleq glacier in western Greenland from 4 August 2021 to the 6th, covering approximately 85 km<sup>2</sup>, with the furthest distance being 13.22 km away from the coastline. Benefited by the meandering coastline, 11 roving stations roughly uniformly distributed on bare rock were surveyed with the RTK technique. PPK-geotagged images were processed in Agisoft Metashape Professional to derive the DSMs, utilizing eight different configurations of GCP distributions that gradually extended longitudinally (along the glacier flow direction) to the upper part of the glacier. The accuracy of DSMs was evaluated by referring to the validation points (VPs) that were not employed in the Bundle Block Adjustment (BBA). The results indicated that the RMSE values of the easting, northing, and height of the reconstruction model georeferenced by only PPK geotagging (no GCPs applied) were 0.038 m, 0.031 m, and 0.146 m, respectively. Applying four GCPs located at one side of the region but with both longitudinal and lateral distribution improved the RMSE values in easting, northing, and vertical to 0.037 m, 0.031 m, and 0.081 m, respectively, and these values were stable even when distributing four GCPs evenly or when increasing the number of GCPs to eleven. Moreover, the cross-validation with ICESat-2 and ArcticDEM performed only at an off-glacier region also suggested that vertical accuracy shows significant improvements for every configuration of GCPs compared to the reconstruction model optimized only by PPK, but such improvements were not obvious if the number of GCPs exceeded four. Moreover, no elevation ramps appeared in the UAV DSM, even for the GCP configuration with only two GCPs distributed at the terminus. Therefore, combining PPK with only a few GCPs but distributing in both directions of the surveying region can offer a viable solution for obtaining glacier DSMs at the coastline with decimeter-level accuracy. |
| format | Article |
| id | doaj-art-b45da0847fdf43149b58a3a9dd8ec9d7 |
| institution | DOAJ |
| issn | 2072-4292 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-b45da0847fdf43149b58a3a9dd8ec9d72025-08-20T02:49:56ZengMDPI AGRemote Sensing2072-42922024-10-011621393410.3390/rs16213934Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, GreenlandHaiyan Zhao0Gang Li1Zhuoqi Chen2Shuhang Zhang3Baogang Zhang4Xiao Cheng5School of Geospatial Engineering and Science, Sun Yat-sen University, and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519082, ChinaSchool of Geospatial Engineering and Science, Sun Yat-sen University, and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519082, ChinaSchool of Geospatial Engineering and Science, Sun Yat-sen University, and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519082, ChinaSchool of Geospatial Engineering and Science, Sun Yat-sen University, and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519082, ChinaFaculty of Geographical Sciences, Beijing Normal University, Beijing 100875, ChinaSchool of Geospatial Engineering and Science, Sun Yat-sen University, and Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai 519082, ChinaReal-Time/Post-Processing Kinematic (RTK/PPK) technology has been widely applied in Unmanned Aerial Vehicle (UAV) photogrammetry in glaciological research. Considering that ground control points (GCPs) cannot be set on glaciers, evaluating the impacts of one-sided distribution is essential. In this study, 8571 images were captured at Sermeq Avannarleq glacier in western Greenland from 4 August 2021 to the 6th, covering approximately 85 km<sup>2</sup>, with the furthest distance being 13.22 km away from the coastline. Benefited by the meandering coastline, 11 roving stations roughly uniformly distributed on bare rock were surveyed with the RTK technique. PPK-geotagged images were processed in Agisoft Metashape Professional to derive the DSMs, utilizing eight different configurations of GCP distributions that gradually extended longitudinally (along the glacier flow direction) to the upper part of the glacier. The accuracy of DSMs was evaluated by referring to the validation points (VPs) that were not employed in the Bundle Block Adjustment (BBA). The results indicated that the RMSE values of the easting, northing, and height of the reconstruction model georeferenced by only PPK geotagging (no GCPs applied) were 0.038 m, 0.031 m, and 0.146 m, respectively. Applying four GCPs located at one side of the region but with both longitudinal and lateral distribution improved the RMSE values in easting, northing, and vertical to 0.037 m, 0.031 m, and 0.081 m, respectively, and these values were stable even when distributing four GCPs evenly or when increasing the number of GCPs to eleven. Moreover, the cross-validation with ICESat-2 and ArcticDEM performed only at an off-glacier region also suggested that vertical accuracy shows significant improvements for every configuration of GCPs compared to the reconstruction model optimized only by PPK, but such improvements were not obvious if the number of GCPs exceeded four. Moreover, no elevation ramps appeared in the UAV DSM, even for the GCP configuration with only two GCPs distributed at the terminus. Therefore, combining PPK with only a few GCPs but distributing in both directions of the surveying region can offer a viable solution for obtaining glacier DSMs at the coastline with decimeter-level accuracy.https://www.mdpi.com/2072-4292/16/21/3934UAVPPKaccuracy assessmentground control pointsGrIS |
| spellingShingle | Haiyan Zhao Gang Li Zhuoqi Chen Shuhang Zhang Baogang Zhang Xiao Cheng Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland Remote Sensing UAV PPK accuracy assessment ground control points GrIS |
| title | Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland |
| title_full | Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland |
| title_fullStr | Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland |
| title_full_unstemmed | Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland |
| title_short | Impacts of GCP Distributions on UAV-PPK Photogrammetry at Sermeq Avannarleq Glacier, Greenland |
| title_sort | impacts of gcp distributions on uav ppk photogrammetry at sermeq avannarleq glacier greenland |
| topic | UAV PPK accuracy assessment ground control points GrIS |
| url | https://www.mdpi.com/2072-4292/16/21/3934 |
| work_keys_str_mv | AT haiyanzhao impactsofgcpdistributionsonuavppkphotogrammetryatsermeqavannarleqglaciergreenland AT gangli impactsofgcpdistributionsonuavppkphotogrammetryatsermeqavannarleqglaciergreenland AT zhuoqichen impactsofgcpdistributionsonuavppkphotogrammetryatsermeqavannarleqglaciergreenland AT shuhangzhang impactsofgcpdistributionsonuavppkphotogrammetryatsermeqavannarleqglaciergreenland AT baogangzhang impactsofgcpdistributionsonuavppkphotogrammetryatsermeqavannarleqglaciergreenland AT xiaocheng impactsofgcpdistributionsonuavppkphotogrammetryatsermeqavannarleqglaciergreenland |