 |
| Titel |
Ground-based multi-view photogrammetry for the monitoring of landslide deformation and erosion |
| VerfasserIn |
André Stumpf, Jean-Philippe Malet, Pascal Allemand, Marc Pierrot-Deseilligny, Grzegorz Skupinski, Christophe Delacourt |
| Konferenz |
EGU General Assembly 2015
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 17 (2015) |
| Datensatznummer |
250109400
|
| Publikation (Nr.) |
 EGU/EGU2015-9305.pdf |
|
|
|
|
|
| Zusammenfassung |
| Recent advances in multi-view photogrammetry have resulted in a new class of algorithms
and software tools for more automated surface reconstruction. These new techniques have a
great potential to provide topographic information for geoscience applications at
significantly lower costs than classical topographic and laser scanning surveys.
Based on several open-source libraries for multi-view stereo-photogrammetry and
Structure-from-Motion, we investigate the accuracy that can be obtained from different
processing pipelines for the 3D surface reconstruc- tion of landslides and the detection
of changes over time. Two different algorithms for point-cloud comparison are
tested and the accuracy of the resulting models is assessed against terrestrial and
airborne LiDAR point clouds. Change detection over a period of more than two
years allows a detailed assessment of the seasonal dynamics of the landslide; the
possibility to estimate sediment volumes, as well as the quantification of the 3D
displacement at most active parts of the landslide. Compared to LiDAR point clouds,
the root-mean squared error of the photogrammetric point clouds did generally
not exceed 0.2 m for the reconstruction of the entire landslide and 0.06 m for the
reconstruction of the main scarp. We show that at the slope scale terrestrial multi-view
photogrammetry is sufficiently accurate to detect surface changes in the range of decimeters.
Thus, the technique currently remains less precise than terrestrial laser scanning or
differential satellite positioning systems but provides spatially distributed information
at significant lower costs and is, therefore, valuable for many practical landslide
investigations. Algorithm parameters and the image acquisition protocols are found to have
important impacts on the quality of the results and are discussed in detail. Our findings
suggest that a relative precision of 1:500 and better is possible. The results of the
change detection show a strong seasonality of the landslide activity and point towards
an increased sediment delivery and transport in spring; especially in years with
intensive and prolonged winters. The quantified volume budgets lead to an average
sediment yield of 426.9 ± 43.9 kg m-2 yr-1 for the monitoring period (2.5 years) which
shows that the landslide scarp must be considered as an important point-source of
loose sediments and a significant term in sediment budgets at the catchment scale. |
|
|
|
|
|
|