2020
DOI: 10.1139/juvs-2019-0006
|View full text |Cite
|
Sign up to set email alerts
|

Evaluation of SfM for surface characterization of a snow-covered glacier through comparison with aerial lidar

Abstract: The combined use of unmanned aerial vehicles (UAVs) and structure-from-motion (SfM) is rapidly growing as a cost-effective alternative to airborne laser scanning (lidar) for reconstructing glacier surfaces. Here we present a thorough analysis of the precision and accuracy of a photogrammetric point cloud (PPC) constructed through SfM from UAV-acquired imagery over the spring snow surface at Haig Glacier, Alberta, Canada, the first of its kind in a glaciological setting. An aerial lidar survey conducted concurr… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
8
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 10 publications
(8 citation statements)
references
References 40 publications
0
8
0
Order By: Relevance
“…In order to assess the accuracy of the setup in the present study, seven ground control points (deployed for the UAV-SfM surveys described in 3.5) were used to compare to the processed UAV-LS data. Due to the limited sample size of GCPs a comparison of the median and Normalised Median Absolute Deviation (NMAD) errors are reported for each day, as well as the median, NMAD, 68.3%, and 95% confidence intervals for all the data, as outlined by Höhle and Höhle (2009) and Bash et al (2020), to better analyse the non-normal distribution of error.…”
Section: Data Quality Assessmentmentioning
confidence: 99%
See 1 more Smart Citation
“…In order to assess the accuracy of the setup in the present study, seven ground control points (deployed for the UAV-SfM surveys described in 3.5) were used to compare to the processed UAV-LS data. Due to the limited sample size of GCPs a comparison of the median and Normalised Median Absolute Deviation (NMAD) errors are reported for each day, as well as the median, NMAD, 68.3%, and 95% confidence intervals for all the data, as outlined by Höhle and Höhle (2009) and Bash et al (2020), to better analyse the non-normal distribution of error.…”
Section: Data Quality Assessmentmentioning
confidence: 99%
“…Each image set was then imported into Agisoft Metashape for image alignment, with a sparse point cloud first generated. Dense point clouds were then constructed using the "aggressive" depth filtering setting, as is common in glacial research (e.g., Tonkin et al, 2014;Jouvet et al, 2019;Bash et al, 2020), before producing DEMs for each survey day. No subsequent mesh or dense cloud smoothing was performed.…”
Section: Uav-sfm (Field Surveys and Processing)mentioning
confidence: 99%
“…While the dominant GSD was ~1.8-2.6 cm, the author reported an RMSE of 0.3-0.4 m; however, that study also found that that the SfM accuracy varied by different surface features, with vegetation, water, and other textures causing more errors. Bash et al [73] analyzed the precision and accuracy of a UAS SfM point cloud versus airborne LiDAR data over a spring snow surface at Haig Glacier; with a GSD of 2.4 cm, most of the errors were within a range of −0.049 ± 0.111 m . Our M3C2 distance between UAS-based LiDAR and SfM point clouds resulted in distance RMSEs of 0.05-0.18 m, which seemed to be slightly better than those from previous studies.…”
Section: Comparison Of the Sampled Ch And Rwmentioning
confidence: 99%
“…However, in densely vegetated areas (dense forest, bushes, or high grass), SfM-MVS can only provide points of the uppermost surface from the sensor's viewpoint, resulting in a surface model rather than a terrain model that is more easily derived from laser scanning data [4,12]. Finally, SfM has been applied at different scales, landscapes, and landforms such as volcanoes and lava movements, large planar regions, glaciers, rock glaciers, badlands, sinkholes, landslides, rivers, burned and aeolian landscapes, as well as at laboratory experiments to detect and analyze surface changes (e.g., [2,[17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33]).…”
Section: Introductionmentioning
confidence: 99%