Development of a computer program from photogrammetry for assisting Q-system rating

Abstract

A computer program to assist the Q-system rating of jointed rock mass is herein developed based on the photogrammetric analysis of point clouds. Three major rating terms, namely, rock quality designation (RQD), joint set number (Jn), and joint roughness number (Jr), are evaluated using predefined algorithms in the computer program without the input of an engineering expert. The random sample consensus algorithm is adopted to identify major outcrop joint surfaces. Thereafter, the major joint surface is determined to calculate the weighted joint density (wJd) and its corresponding RQD. In addition, a spherical K-means clustering is utilized to obtain the number of dominant joint sets and further convert it to the Jn value. Moreover, the a/L method is adopted to calculate the joint roughness coefficient (JRC) of those facets and convert it to Jr. The analysis results agree with manual Q-system rating results, revealing that the proposed computer program is practicable. Furthermore, we evaluate the limitation of the proposed computer program as applied to interlayered sedimentary rocks. Finally, we found that doubling the standard deviation of the distance between fitted facets to the point cloud is a reasonable estimate of the JRC20 and Jr of rock joints.