Horizontal displacement monitoring in mining areas using UAV data from photogrammetry and from laser scanning

Abstract

In monitoring surface deformation related to underground mining activities, analyzing both vertical (dU) and horizontal (dN, dE) displacements is essential. Most studies focus on 1-dimensional vertical movements, overlooking horizontal displacements, which are crucial for understanding the full 3-dimensional deformation pattern caused, which reduces the effectiveness of risk assessment. This study explores four approaches for analyzing horizontal displacements using image-based, point cloud-based, and surface-based methods, utilizing photogrammetric and laser scanning data from unmanned aerial vehicles (UAVs). The methods were evaluated in terms of accuracy, usability, and limitations. The accuracy analysis was conducted using Global Navigation Satellite Systems (GNSS) as reference data. The results showed that the highest accuracy was achieved with UAV photogrammetry, using automatic analysis of corresponding points from raw images (RMSE: 0.030 m, 0.020 m, 0.032 m for the dN, dE, and dU displacement components, and 0.031 m for the horizontal displacement magnitude). However, this method is most effective in urban areas with consistent lighting and stable objects. For point clouds and digital surface models (DSMs), the results achieved accuracies of 0.029 m, 0.027 m, and 0.047 m for the dN, dE, and dU displacement components, respectively, and 0.031 m for the horizontal displacement magnitude. Despite this, point clouds and DSMs are recommended for displacement monitoring, as they provide more reliable data in varying lighting conditions and natural environmental changes, offering a more consistent and reliable approach than RGB imagery.