Testing and Evaluation of the Deformation Effects Developed By Landslides in the Laboratory With Image Processing Techniques

Abstract

Landslides are one of the most frequently encountered natural disasters in Türkiye and the world, which can cause significant loss of life and property. Nowadays, various methods such as inclinometers, tiltmeters, extensometers, fiber optic methods, and ground-based surface measurement systems (i.e., LIDAR, electro-optics, total station) are used to monitor sliding mass with potential stability risks. In recent years, image processing techniques have become popular; however, their use in landslide investigations has not yet expanded to the desired level. With the development of image processing techniques, image recordings are used in many sectors for analysis during or after the studies. By this way, real-time data collection and rapid response capabilities can be increased to manage possible disaster situations effectively. This thesis uses image processing methods to detect the deformation movement on the sliding surface caused by any triggering mechanism (i.e., excessive rainfall, earthquake, human factors, sudden snowmelt), regardless of lithological unit and failure types. For this purpose, the deformation movement on an artificial landslide simulator created on a shaking table operating as a triggering mechanism in the laboratory environment was analyzed. The tilt sensor was integrated into the landslide simulator in the vertical direction, and laser measurements were performed to evaluate the accuracy of the results obtained by image processing. This study is a prototype that analyses mass movements from different perspectives in a laboratory environment. It offers new perspectives for monitoring landslides and, accordingly, for natural disaster management and risk reduction strategies.