A rigorous cylinder-based self-calibration approach for terrestrial laser scanners

Existing self-calibration methods for terrestrial laser scanners are predominantly point-based and plane-based. In this paper, we present a new cylinder-based self-calibration method with its variants for several scanners having different architectures and scanning mechanisms. The method not only increases the flexibility of in situ self-calibration, but also its rigor because of reduced functional dependencies between adjustment parameters. Based on the analysis of linear dependencies between columns of the design matrices for both the cylindrical and planar models, it is shown that using the vertical cylindrical model is advantageous over using the planar model as some high linear dependencies can be avoided. The proposed method and its variants were first applied to two simulated datasets, to compare their effectiveness, and then to three real datasets captured by three different types of scanners are presented: a Faro Focus 3D (a phase-based panoramic scanner); a Velodyne HDL-32E (a pulse-based multi spinning beam scanner); and a Leica ScanStation C10 (a dual operating-mode scanner). The experimental results show that the proposed method can properly estimate the additional parameters with high precision. More importantly, no high correlations were found between the additional parameters and other parameters when the network configuration is strong. The overall results indicate that the proposed calibration method is rigorous and flexible.