Abstract
The horizontal positional accuracy of LiDAR data is important for quality control monitoring, however it is often difficult to estimate. This difficulty is due in part to the problems observing ground truth features in the point cloud and determining the coordinates of surveyed marks on these features. Purpose designed targets can be deployed to provide ground truth, but this can be inconvenient. Methods have been proposed that use existing ground features, but the shapes and orientations of these markings tend to be site specific. In this paper the idea of using roundabouts as ground truth features is presented. Roundabouts are common in many urban environments and have useful properties for accuracy estimation. These properties can be used to find roundabout centres and then compared with survey mark coordinates to estimate the horizontal accuracy of the data; with an uncertainty of around 0.15m. The methods and algorithms used are presented, results of testing provided along with thoughts for further research.
Highlights
The ”Julia” programming language is used for algorithm development, data manipulation and plotting.Horizontal accuracy testing for LiDAR data has challenges due in part to the discrete nature of LiDAR point clouds and the often remote and large survey extent
An edge of a building might be between two flight lines in the point cloud
The cropped point cloud is ordered by time and scan line as in the intensity gradient method
Summary
The ”Julia” programming language is used for algorithm development, data manipulation and plotting.Horizontal accuracy testing for LiDAR data has challenges due in part to the discrete nature of LiDAR point clouds and the often remote and large survey extent. The method could potentially most cases, include just the roundabout, adjacent road surface and provide estimations of horizontal accuracy if terrestrial data was vehicles. They the gradient method only, is to average the Easting and Northing estimate the accuracy of computed target coordinates based on error coordinates of all points contained within the circumference
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call