Abstract
It is important to monitor the stability of open pit slopes for worker, public, and environmental safety. Depending on the personnel and equipment resources available on site, some monitoring methods may be preferred over others. The Faro Mine Complex (FMC), a site currently in care and maintenance, has three inactive open pits. Since 2016, a photogrammetry network has been used as a cost effective, easy to use, and relatively low maintenance solution for pit slope monitoring at the FMC. This system was set up with the goal of detecting slope displacements of 20 cm or greater, which is considered sufficient for large-scale slope stability and pit crest regression monitoring at the site. To detect slope displacements, 3-D point clouds generated from two different dates of photograph data collection are compared. To accurately measure differences (i.e., slope displacements) between the two point clouds, they must first be aligned. This alignment is routinely done by use of an iterative closest point algorithm, followed by a cloud-to-cloud nearest neighbour distance calculation where the point clouds are aligned using the interpreted stable areas of the slope, similar to the process routinely used for laser scan data processing (Abellán et al. 2010; Lague et al. 2013; Kromer et al. 2015). Use of this conventional approach has typically required multiple iterations to achieve point cloud alignment and has resulted in variable detection limits. An alternative approach where point clouds are automatically aligned by the photogrammetry software matching thousands of common points on both sets of photographs from the two data sets has recently been implemented for monitoring pit slope stability at the FMC. This new approach has improved the average displacement detection limit to 13 cm and has greatly reduced model processing time while improving reliability.