Abstract
Landslides are major and constantly changing threats to urban landscapes and infrastructure. It is essential to detect and capture landslide changes regularly. Traditional methods for monitoring landslides are time-consuming, costly, dangerous, and the quality and quantity of the data is sometimes unable to meet the necessary requirements of geotechnical projects. This motivates the development of more automatic and efficient remote sensing approaches for landslide progression evaluation. Automatic change detection involving low-altitude unmanned aerial vehicle image-based point clouds, although proven, is relatively unexplored, and little research has been done in terms of accounting for volumetric changes. In this study, a methodology for automatically deriving change displacement rates, in a horizontal direction based on comparisons between extracted landslide scarps from multiple time periods, has been developed. Compared with the iterative closest projected point (ICPP) registration method, the developed method takes full advantage of automated geometric measuring, leading to fast processing. The proposed approach easily processes a large number of images from different epochs and enables the creation of registered image-based point clouds without the use of extensive ground control point information or further processing such as interpretation and image correlation. The produced results are promising for use in the field of landslide research.
Highlights
Changes and displacements are fundamental indications of the Earth’s surface’s mass movements, such as landslides, soil creep, and rock slides, which are caused by either human activity or natural processes
Image acquisition was performed during camera in operating medium field-of-view mode, thefield flight lines wereover captured at a rate five the two flights differentat directions (Figure 3) for each operation the study area.of With camera operating at medium field-of-view mode, the flight lines were captured at a rate of five frames per second with an altitude of roughly 30 m and at a speed of 5 m/s, resulting in a ground sample distance (GSD)
Image-based point clouds are generated for each epoch separately (i.e., structure from motion (SfM) and semi-global matching (SGM) for each epoch were processed individually) to implement the iterative closest projected point (ICPP) registration approach
Summary
Changes and displacements are fundamental indications of the Earth’s surface’s mass movements, such as landslides, soil creep, and rock slides, which are caused by either human activity or natural processes. The application of 3D point correspondence using a point-to-patch comparison for detecting volumetric surface changes is presented as well as the change displacement rate in the horizontal direction based on comparisons extracted from landslide scarps from image-based point clouds generated from multiple epochs. This area has a semi-arid climate with an average maximum temperature of 12.3 °C and an average minimum temperature of −1.1 °C. Non-cultivated parts of the area are covered by the typical prairie flora
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
