COMPARING TERRESTRIAL LIDAR WITH STRUCTURE FROM MOTION FOR DISCONTINUITY ORIENTATION MEASUREMENT

Orientation of discontinuities are one of the most important aspects of rock mass characterization. Transit compasses are traditionally used to take discontinuity orientations (strike, dip direction, dip) data in the field. The use of Terrestrial LiDAR (TLS) for discontinuity orientation data has become more common as it is more efficient and accurate than the traditional methods. TLS is a surveying method involving the generation of a point cloud from pulsed laser shot by the TLS scanner and reflected back from solid surfaces generating millions of XYZ points known as point cloud. 3D surfaces can be created from the point cloud on which discontinuities can be identified and measured. Recently structure from motion (SfM) has emerged as an alternative method for creating point clouds from stitching of overlapping photographs, which can be turned into a 3D models. Data collection for SfM only requires the use of a smartphone or compact digital camera instead of an expensive instrument such as the TLS. Data processing required a relatively cheap software, Agisoft. Once the 3D models are generated, discontinuities can be identified and their orientations measured. The purpose of this project is to compare SfM to TLS with respect to extracting orientation of discontinuities. Through preliminary testing, we were able to determine that photo density, photo resolution and light condition were the main factors in quality of 3D models generated from SfM. To compare SfM with LiDar a quarry, Frazier Quarry in Harrisonburg, VA was chosen as a test site. Photo density, photo resolution and light condition are the three variables that will be tested for accuracy of discontinuity orientation measurements extracted from SfM. Both SfM and TLS will be compared in terms data collection time, processing time, accuracy of measurements, and the software/hardware costs.