Spacing and block volume estimation in discontinuous rock masses using image processing technique: a case study

Mohammad Azarafza¹. Mustafa K. Koçkar¹. Lohrasb Faramarzi²

¹Department of Civil Engineering, Hacettepe University, Ankara, Turkey

²Department of Mining Engineering, Isfahan University of Technology, Isfahan, Iran

Doi:10.1007/s12665-021-09768-3

Abstract
Application of the image processing techniques (IPT) to identify rock mass geometry provides more fast information about discontinuity properties used in geo-engineering characteristics. In this regard, the field survey can be improved using IPT. This study has utilized the IPT to identify the discontinuity and block volume characteristics in a discontinuous rock mass. For this purpose, a visual evaluation of the rock mass outcrop with discontinuities from a road slope cut located in the South Pars Special Zone, Assalouyeh, Iran, was considered. A three-step IPT analysis (i.e. pre-processing, main processing, and post-processing) was conducted to extract the features through the Python programming language. Regarding the IPT methodology, the studied rock mass characteristics consist of four major discontinuity sets and rock block volumes between the intersections of the discontinuities, as confirmed with a scan-line field survey. The evaluated data indicated that the maximum, minimum, and average block volumes processed by the IPT were 1.068, 0.479, and 1.055 m3, and their field measurement results were 1.092, 0.479, and 1.065 m3, respectively. Additionally, the orientations of the estimated discontinuity properties and their spacings determined by IPT for the rock mass ranging between 32 and 69.9° and 0.5 and 2.18 m, respectively. Similarly, the orientations of the field measurement results were also obtained between 33 and 71° and 0.58 and 2.25 m, respectively. The results of the IPT and the field survey were close, which revealed that the IPT is a reliable method for determining discontinuity spacing and rock block volume along large cut slopes. This approach provided rapid data processing with spatial extensions in a short period, making it possible to achieve accurate results in discontinuity network characteristics.

Keywords: Image processing technique. Artificial intelligence. Python programming language. Rock engineering