Elementary Analysis of the Mechanism of Xishan Landslide Based on Pixel Tracking on VHR Images

Authors

  • Beiqi Shi Shanghai Institute of Tourism, Shanghai Normal University, Shanghai
  • Chun Liu TongJi University, Shanghai
  • Hangbin Wu TongJi University, Shanghai
  • Ping Lu TongJi University, Shanghai

DOI:

https://doi.org/10.15273/ijge.2016.02.007

Keywords:

landslide, deformation monitoring, Very High-Resolution Imagery, digital image correlation, strain analysis

Abstract

The geometric and kinematic characterizations of landslides are key factors for understanding the mechanisms of landslide movement. Recent developments in remote sensing techniques provide powerful tools for landslide deformation studies. The objective of this paper is to present an efficient method for monitoring the horizontal ground displacement and strain field of active landslides from space-borne optical images, which were analyzed with a digital image correlation technique. This method was evaluated on a series of images acquired on the Xishan landslide in China over the period between 2010 and 2013. These imageries included ZY-3 and WorldView-II images. These were Very High Resolution (VHR) satellite imageries and were orthorectified with a high resolution DEM interpolated from Unmanned Aerial Vehicle (UAV). By using sub-pixel correlation on before- and after-event orthoimages, it was possible to compute the displacement field with high planimetric resolution. Furthermore, a strain analysis was performed to characterize the deformation and displacement regime of the landslide. Validation of the computed strain field analyzed the relationship between the deformation and the faults detected in the field. This approach can facilitate displacement measurements. It could, therefore, offer promising potential for operational applications, for instance, early warning  of of geo-hazards.

Downloads

Issue

Section

Articles