Abstract

LiDAR in City Modelling - Different Approaches and Deliveries
Airborne and mobile LiDAR surveying are different methods to make city models for different purposes. The key-issues in both method are cheap data collection, automatic data processing, easiness to update models and a possibility to monitor the changes in the environment automatically. Airborne LiDAR suits the best to survey large areas whereas mobile LiDAR systems survey corridors and produce models of street and building facades. The quality of airborne models is around constant: the same point density and accuracy on the whole area. This information can also be used to fix the position of mobile LIDAR in cases, where control measurements are impossible. The point density is crucial to detect buildings and their details. The LoD level of buildings is 1 or 2 To vectorize mobile LiDAR data is an endless work and does not lead to good results. An advanced point coloring is a new solution to the problem. The method selects the best raster from images to color corresponding point, removes moving objects like cars, colorize thin objects like lamp poles and wires by intensity or uses mixed artificial colors to trees. In the process a careful data classification is essential. Luckily many of the processes, which are still manual, can be atomized to decrease manual work. Exact LoD3 level models as point clouds are easy to move between different systems. Besides using point clouds one can also produce hybrid models by vectorizing selected objects partly or completely. The presentation shows an example model done in 2014 in Helsinki city.