Optimization of point clouds for 3D bas-relief modeling

Abstract

Terrestrial Laser Scanning (TLS) is an ideal, non-invasive remote sensing technology for all measurements related to examining the condition of building walls. Based on the point clouds, users can carry out dimensioning and 3D modeling of bas-reliefs. TLS gives the opportunity to obtain point clouds with high resolution, which allows for accurate mapping of objects, but on the other hand is troublesome during data processing. Therefore, this paper presents a methodology of TLS point cloud optimization based on a modified Optimum Dataset method (OptD v.A). OptD v.A allows the number of observations to be reduced. The reduction does not happen in a random manner, the points are examined for their usefulness and relevance. The research was carried out using real walls with bas-reliefs. The tests showed that the OptD v.A method did not disturb the nature of the tested objects. Optimized TLS point clouds were used for 3D modeling.