Rock Mass Surface Roughness Characterization Using Image Analysis Technique

The joint roughness coefficient (JRC) is very important to determine the shear strength of the rock, for example, as one of the inputs for the Barton-Bandis model. Conventionally, the Barton comb profilometer is widely used in the field but some issues related to accessibility, labor-intensive, a...

Full description

Bibliographic Details
Main Author: Adnan, Raja Asyraf Azizan Raja
Format: Monograph
Language:English
Published: Universiti Sains Malaysia 2021
Subjects:
Online Access:http://eprints.usm.my/57295/
http://eprints.usm.my/57295/1/Rock%20Mass%20Surface%20Roughness%20Characterization%20Using%20Image%20Analysis%20Technique.pdf
Description
Summary:The joint roughness coefficient (JRC) is very important to determine the shear strength of the rock, for example, as one of the inputs for the Barton-Bandis model. Conventionally, the Barton comb profilometer is widely used in the field but some issues related to accessibility, labor-intensive, and time-consuming. To tackle these problems, this study aims to evaluate the photogrammetry technique in producing reliable JRC measurements. To achieve this, a set of JRC replica of the rock mass is produced to determine the JRC readings. In addition, a drone is used to take photos of the JRC model with high quality as Unmanned Aerial Vehicle (UAV) photogrammetry method. The reliability of such measures depends on some parameters such as the number of images, image quality, and the number of point clouds. The digitalization of the JRC model will take place to create a 3D model using photogrammetry. The JRC measurement results are compared with the manual Barton comb profilometer method. Then, an actual rock mass is used to verify the photogrammetry technique. As result, the JRC of the 3D model can be produced by using the image analysis technique. The ultra-high quality has the most accurate measurement as actual length with zero percent error compared to actual measurement using Barton comb. As for low, medium, and high quality, the error was 15.54%, 9.46%, 2.7% respectively to the actual. However, the medium quality is the most efficient way since it can produce the reliable JRC measurement within a short period and can practically use for fieldwork.