Targeted Rock Slope Assessment Using Voxels and Object-Oriented Classification

Reality capture technologies, also known as close-range sensing, have been increasingly popular within the field of engineering geology and particularly rock slope management. Such technologies provide accurate high-resolution n-dimensional spatial representations our physical world, 3D point clouds, that are mainly used for visualization monitoring purposes. To extract knowledge from clouds inform decision-making management systems, semantic injection through automated processes is necessary. In this paper, we propose a model utilizes segmentation procedure which delivers segments ready to classify be retained or rejected according complementary knowledge-based filter criteria. First, relevant voxel-based features based on local dimensionality, orientation, topology partition them in an assembly homogenous segments. Subsequently, build decision tree geometrical, topological, contextual information enables classification multi-hazard railway section British Columbia, Canada into classes involved landslide risk Finally, approach compared machine learning integrating recent featuring strategies with limited training data (which usually case). This alternative approaches reduces substantially size complexity provides adaptable framework tailored systems leveraging semantics.