Tag Archives: feature extraction

New publication in Computer and Geosciences

A new approach for semi-automatic rock mass joints recognition from 3D point clouds

Riquelme, A., Abellan, A., Tomás, R., Jaboyedoff, M (2014). A new approach for semi-automatic rock mass joints recognition from 3D point clouds. Computers & Geosciences, 68(0), pp.38–52. DOI: 10.1016/j.cageo.2014.03.014

In this paper we present a semi-authomatic methodology for the identification and analysis of flat surfaces outcropping in rocky slopes from 3D point clouds. The method is based on a three steps process: (i) to carry out a nearest neighbour points coplanarity test and to computate the normal vector; (ii) to find the principal orientation families by Kernel Density Estimation; and (iii) to identify and extract clusters of points belonging to the same discontinuity.

Different sources of information – synthetic and 3D scanned data – were employed in this study, being raw source files and obtained results freely provided aiming to a more reproducible research.

Discontinuity extraction
Example of clusters identification. (a) One colour per discontinuity set with all clusters labelled; from (b) to (f) family sets representation using one colour per cluster (Riquelme et al., 2014).