Bias Correction for View-limited Lidar Scanning of Rock Outcrops for Structural Characterization

Lidar is a remote sensing technology that uses time-of-flight and line-of-sight to calculate the accurate locations of physical objects in a known space (the known space is in relation to the scanner). The resultant point-cloud data can be used to virtually identify and measure geomechanical data such as joint set orientations, spacing and roughness. The line-of-sight property of static Lidar scanners results in occluded (hidden) zones in the point-cloud and significant quantifiable bias when analyzing the data generated from a single scanning location. While the use of multiple scanning locations and orientations, with merging of aligned (registered) scans, is recommended, practical limitations often limit setup to a single location or a consistent orientation with respect to the slope and rock structure. Such setups require correction for measurement bias. Recent advancements in Lidar scanning and processing technology have facilitated the routine use of Lidar data for geotechnical investigation. Current developments in static scanning have lead to large datasets and generated the need for automated bias correction methods. In addition to the traditional bias correction due to outcrop or scanline orientation, this paper presents a methodology for correction of measurement bias due to the orientation of a discrete discontinuity surface with respect to the line-of-sight of the Lidar scanner and for occlusion. Bias can be mathematically minimized from the analyzed discontinuity orientation data.