The study of the mechanical behaviour of fractured rock masses should always be based on accurate rock surveys of the geological structure. Traditionally surveys are performed manually by using compass and directly accessing to the rock mass: this is often dangerous or unpractical while the samples size is small, i.e. not optimal to characterize the rock mass completely. In this paper, alternative techniques will be discussed. Using photogrammetry or a laser scanner, an accurate 3D digital model (DSM) of the rock surface is generated. Discontinuity orientation and position on the rock face are derived from the DSM in order to perform a deterministic reconstruction of the rock mass and the determination of the rock blocks lying on the slope. Different levels of automation in discontinuity planes detection were implemented: interactive identification of single planes and interactive selection of macro-areas with automatic segmentation of planes. Segmentation results are organized in order to be directly used for the deterministic reconstructions of the rock masses and the study of the rock mass stability conditions with the key block method and the Distinct Element Method in a 3 dimensional field. These alternative methodologies have been validated trough the comparison with traditional surveys in two different pilot sites (Arnad, North West Italy and Le Trappistes, Swiss).
The study of the stability condition of a rock mass is usually based on a geo structural survey of the existing discontinuities. Surveys are devoted to a systematic and quantitative description of rock discontinuities and can be executed on rock faces, natural or artificial like excavations, or on boreholes.
Traditionally surveys are performed by using a compass directly applied on the rock face. In several cases, this can be difficult because rock masses: cannot be easily accessed or the size of rock faces is so large that the data acquisition may be long and expensive when done manually; are so heavily fractured and not supported that it can be dangerous to climb the slope or even stand at its base.
Sometimes the on site interpretation of the data is not easy; a better understanding of the rock structure can be achieved later in the office, when the data have been properly organized and compared; alternative methods may also allow the possibility to perform additional measurements without coming back to the field.
Traditional surveys are carried on scan-lines or, in a more detailed way, through observation windows (figure 1). Nevertheless, they do not allow, in several cases, to identify important fracturation aspects that are needed to carry on stability analysis.