QUANTITATIVE EVALUATION AND QUALITY CONTROL OF COMMERCIALLY ADOPTED TRADITIONAL AND MODERN LIDAR SYSTEM CALIBRATION TECHNIQUES

Abstract

LiDAR systems have been widely adopted for the collection of topographic data. By integrating the information gathered by navigation sensors (GPS/INS) and a laser ranging/scanning unit, LiDAR systems can directly provide the 3D coordinates of a surface at a high density. In the past decade, LiDAR technology has undergone significant improvements in performance (e.g., higher pulse repetition frequencies, higher operational altitudes) and data processing/post-processing methodologies. Major advances in the data processing include more robust methodologies for the system calibration. Implemented traditional calibration by the service providers are based on iterative sequential estimation of the system parameters that require manual adjustment and time-intensive interaction of a trained operator. In the past few years, automated and more accurate methodologies have become commercially available and have been currently in use by some data providers. In this paper, traditional and modern LiDAR system calibration procedures are evaluated and compared. For that purpose, a practical quality control procedure is used. The underlying concept of the quality control procedure is that in the absence of biases in the system parameters (i.e., for a properly calibrated system), conjugate surface elements in overlapping strips should coincide with each other as well as possible. Incompatibilities between conjugate surface elements in overlapping strips can be used to evaluate the quality of the calibration process. In addition, the presented quality control procedure can be used for diagnosing the cause of detected incompatibilities. More specifically, the detected incompatibilities can be used for estimating the remaining biases in the system parameters. Another advantage of the introduced quality control procedure is the possibility of its implementation by the end user since it only requires the LiDAR point cloud coordinates as well as a general knowledge of the flight configuration. Experimental results have demonstrated significant improvements in the quality of fit among overlapping LiDAR strips when using modern LiDAR system calibration procedures and the ability of the proposed quality control approach to detect and eliminate remaining biases in the system parameters.