Object-oriented change detection and damage assessment using high-resolution remote sensing images, Tangjiao Landslide, Three Gorges Reservoir, China

Abstract

This paper presents a new approach for tracking land cover changes and assessing damages caused by individual reservoir landslides using high-resolution images. The object-oriented change detection (OOCD) approach has recently become more popular than traditional pixel-oriented methods for high-resolution image analysis. However, few studies have applied the OOCD approach to the land cover change detection (LCCD) and the damage assessments of individual landslide. An OOCD approach was applied to the multi-temporal high-resolution images taken in 2002, 2005, 2010 and 2013 throughout the Tangjiao Landslide in the TGRA. The proposed OOCD approach contains three major steps: land-cover-type analysis, high-resolution image classification using an object-oriented classification method and LCCD via comparing the classified geographic objects in different temporal images. The object-oriented classification results show that the overall classification accuracies of the 2005, 2010 and 2013 images were greater than 92%, with Kappa Index of Agreement values of at least 89%. The IKONOS image taken in 2002 is an exception to both of these values. Land cover change maps suggest that various damages occurred throughout the study area from 2002 to 2013. Large woodland areas were submerged at the front of the study area. In addition, 21,365 m2 buildings and a 300-m-long road were damaged between 2005 and 2010. Furthermore, a large surface crack was observed from 2010 to 2013. The results suggest that the OOCD approach is able to effectively and accurately identify damage characteristics on the individual landslide. The results also show that the instability of Tangjiao Landslide is controlled by the geological conditions and seriously affected by the reservoir water-level fluctuation and seasonal rainfall.