Giovanni Mastrorocco

The use of an unmanned aerial vehicle for fracture mapping within a marble quarry (Carrara, Italy): photogrammetry and discrete fracture network modelling

ABSTRACT This paper describes the use of a drone in collecting data for mapping discontinuities within a marble quarry. A topographic survey was carried out in order to guarantee high spatial accuracy in the exterior orientation of images. Photos were taken close to the slopes and at different angles, depending on the orientation of the quarry walls. This approach was used to overcome the problem of shadow areas and to obtain detailed information on any feature desired. Dense three-dimensional (3D) point clouds obtained through image processing were used to rebuild the quarry geometry. Discontinuities were then mapped deterministically in detail. Joint attitude interpretation was not always possible due to the regular shape of the cut walls; for every discontinuity set we therefore also mapped the uncertainty. This, together with additional fracture characteristics, was used to build 3D discrete fracture network models. Preliminary results reveal the advantage of modern photogrammetric systems in producing detailed orthophotos; the latter allow accurate mapping in areas difficult to access (one of the main limitations of traditional techniques). The results highlight the benefits of integrating photogrammetric data with those collected through classical methods: the resulting knowledge of the site is crucially important in instability analyses involving numerical modelling.

Application of UAV photogrammetry for the multi-temporal estimation of surface extent and volumetric excavation in the Sa Pigada Bianca open-pit mine, Sardinia, Italy

In open-pit mines, monitoring of topographic and volumetric changes through time is found to be of great importance to support excavation stages and to plan rehabilitation strategies. In this work, we describe a geomatic approach to assess changes in surface mine extent and to quantify excavated volume in the Sa Pigada open-pit mine, Sardinia, Italy. We performed two drone-based photogrammetric surveys in 2013 and 2015, and by means of the Structure from Motion (SfM) technique, we obtained related 3D dense point clouds and digital orthophotos. Images were georeferenced thanks to a series of ground control points surveyed with geodetic GPS. Distances between the two clouds were estimated with the recent Multiscale Model to Model Cloud Comparison (M3C2) plug-in included in the CloudCompare open-source software. Starting from cloud-to-cloud distances, we calculated the excavated volume of mineral resources between the two surveys. Results of the M3C2 comparison supported the analysis of the two orthophotos, through which accurate limits of the 2013 and 2015 active mine areas, rehabilitated area and temporary dumps were identified and drawn in a digital map. Results obtained in this study suggest that the applied geomatic techniques are suitable for performing accurate change detection analysis in open-pit environments and represent a valid support for scientists and technicians allowing to monitor with high spatial and temporal resolutions. This approach can be also considered a valid tool to reduce environmental impact from mining.

Fracture mapping in challenging environment: a 3D virtual reality approach combining terrestrial LiDAR and high definition images

The latest technological developments in computer vision allow the creation of georeferenced, non-immersive desktop virtual reality (VR) environments. VR uses a computer to produce a simulated three-dimensional world in which it is possible to interact with objects and derive metric and thematic data. In this context, modern geomatic tools enable the remote acquisition of information that can be used to produce georeferenced high-definition 3D models: these can be used to create a VR in support of rock mass data processing, analysis, and interpretation. Data from laser scanning and high quality images were combined to map deterministically and characterise discontinuities with the aim of creating accurate rock mass models. Discontinuities were compared with data from traditional engineering-geological surveys in order to check the level of accuracy in terms of the attitude of individual joints and sets. The quality of data collected through geomatic surveys and field measurements in two marble quarries of the Apuan Alps (Italy) was very satisfactory. Some fundamental geotechnical indices (e.g. joint roughness, alteration, opening, moisture, and infill) were also included in the VR models. Data were grouped, analysed, and shared in a single repository for VR visualization and stability analysis in order to study the interaction between geology and human activities.

Remote sensing techniques in underground mining: Application of terrestrial LIDAR for stability analysis and optimization of excavation activities

Abstract from 88th Congress of the Italian Geological Society, 2016-09-07 - 2016-09-09, NaplesAbstract from 88th Congress of the Italian Geological Society, 2016-09-07, 2016-09-09, Naplesbook Edited by D. Calcaterra, S. Mazzoli, F.M. Petti, B. Carmina & A. Zuccari doi: 10.3301/ROL.2016.79

3D point cloud analysis for surface roughness measurement: Application of UAV photogrammetry

This paper describes the application of close-range digital photogrammetry through Unmanned Aerial Vehicle (UAV) for rock discontinuity roughness determination. This was possible thanks to the creation of a georeferenced point cloud characterised by 3D information useful for rock mass study. In particular, UAV systems allows producing dense point clouds (with high spatial resolution), suitable to describe the rock surfaces and allowing the surveying of inaccessible outcrops. In fact, the occlusion phenomenon was minimized by means of different acquisition angles. In this context, the computation of the Joint Roughness Coefficient (JRC) was carried out picking several points within the cloud and obtaining rock profiles of the maximum asperity amplitude. Furthermore, we present in the manuscript a rapid and preliminary estimation of the whole rock slope roughness using the freeware CloudCompare code. A 3D roughness model was generated. The JRC results, both manually measured and automatically computed by the code, are comparable. The reliability of the values was confirmed by direct engineering-geological surveys where the surface roughness was calculated using the traditional method of visual comparison to Bartons reference roughness profiles. The paper aims to demonstrate how data derived from UAV processing are, nowadays, so dense and accurate to be able to provide adequate measurements of discontinuity roughness even transcending the problem of the JRC scale effect.