Digital Urban Morphometrics: Automatic Extraction and Assessment of Morphological Properties of Buildings

The aim of this article is to present a method to calculate the morphological properties of the built environment using LiDAR (light-detection and ranging) data, geographic information systems (GIS) data and three-dimensional (3D) models of cities as a source of information. A hybrid approach that takes into account different types of inputs and consequently evaluates the accuracy of each type of used data is presented. This work is intended to give a first response to the lack of comprehensive and accurate procedure that uses LiDAR data in order to automatically derive precise morphological properties, such as volumes and surfaces (façades and roofs) of buildings. The method was tested on two case-study areas in the Geneva region with different characteristics, one in the old town along the Rhone River and the other on the CERN campus. A statistical analysis that compares the results of the computation with the 3D model of the built environment was used to validate the results, complemented by significance statistical tests. Outcomes showed that the proposed method to derive morphological properties can reach high levels of accuracy, thus enhancing the potential uses of LiDAR data for numerous applications, typically for the assessment of the urban environmental quality (UEQ) at the city and district scale, such as the estimation of the potential deployment of renewable energies in the built environment and the determination and monitoring of several urban indicators.