Flight optimization algorithms for aerial LiDAR capture for urban infrastructure model generation

Aerial Light Detection and Ranging (LiDAR) offers the potential to auto-generate detailed, three-dimensional (3D) models of the built environment in urban settings. Auto-generation is needed as manual generation is not economically feasible for large areas, and yet such models would offer distinct advantages for a wide range of applications from improved noise and
pollution prediction to disaster mitigation modeling. Current technology and the dense geometry of urban environments are two major constraints in LiDAR scanning. This paper outlines the difficulties related to effective vertical surface data capture in an urban environment for the purpose of 3D visualization. Further, the traditional point data captured with LiDAR scans is unsuitable for visualization. Therefore, surface reconstruction algorithms must be applied to the data. These algorithms are largely dependent on the uniformity of the resolution in the point data. Principles for geometric optimization of data capture on vertical surfaces, thereby improving resolution uniformity, are presented.