Evaluation of projection model for random point cloud

Recently, point cloud data are acquired by some platforms, such as a terrestrial laser scanner, land-based mobile mapping system (MMS), and airborne LiDAR. These systems can achieve a rapid and massive point cloud data acquisition for road surveying, mapping, structure maintenance, and environment monitoring. However, massive point cloud data require huge processing time in data sharing, visualization and 3D modeling. Therefore, we have proposed a performance improvement of point cloud processing based on point-based rendering approach. Our point-based rendering can select several projection models, such as a spherical, cylindrical, and orthogonal model. Each model has different advantages and disadvantages. Therefore, we proposed a methodology to select a suitable projection model in some point cloud editing works in a road monitoring, structure monitoring, surveying, and indoor mapping. In this paper, we evaluated each projection models through some experiments using terrestrial LiDAR and MMS data.