3D Urban Models from Laser Radar Data

Updated geographic information and 3D models of environments are becoming increasingly important for applications such as city planning, crisis management, visualisation, architecture, and landscaping. The task of producing detailed geospecific terrain databases or 3D city models has traditionally been manual and thus time-consuming, and the availability of appropriate source data has been limited. The use of available sensors such as airborne laser scanners (ALS) and digital cameras provides new opportunities to acquire detailed remote sensing data of the natural environment. This type of data is very suitable as the basis for the construction of high-fidelity 3D virtual environment models. To make use of this new type of data and to develop applications as mentioned above requires new methods for processing 3D sensor data and extracting geographic feature information from the data. In this paper we present recent results from the development of methods for processing ALS data and for transforming the data into geographic information and environment models, with the emphasis on urban environments. The long term objective of our work is the development of new methods for rapid and highly automated extraction of geographic information to support the construction of high-fidelity 3D virtual environment models from remote sensing data. As remote sensing data we use data from current high resolution ALS systems and digital cameras.