New image processing algorithm for geological structure identification of rock slopes based on drone-captured images, A

Abstract

Rock slope failure is a severe phenomenon which may lead to fatalities and injuries in open pit mines and quarries. Many historical cases of rock slope failure show the adverse effects of geological structures on rock slope stability. In this study, a new algorithm based on drone photogrammetry is introduced. First, high resolution images are acquired with a drone to identify geological structures foror a give part of a rock slope. The acquired digital images are then processed in Image J. Next, MATLAB software is used to automatically identify the geological structures and group them according to their geometric properties. In addition, the rockmass conditions of the selected rock slopes are also evaluated. RocData, software is then used to estimate the strength parameters of the rock slope using Hoek-Brown Failure Criterion. DIPS 7.0 was used to plot identified geological structures on a stereonet and to assess the likelihood of different failure modes. To demonstrate the application of the proposed method, three case studies of rock slopes were presented, one a natural slope, and the other two an open pit and a quarry. The results illustrate that drone-based photogrammetry can provide more information, and more geological structures can be identified than from ground-based photogrammetry. The proposed method is useful for geological structures identification on rock slopes, increasing the slope monitoring accuracy, and improving efficiency and safety.