4D city models enable detection of changes in built environment over time
The research team led by Prof. Xiaoxiang ZHU in the Technical University of Munich (TUM) retrieved data from the German TerraSAR-X satellite, the most high-resolution civilian radar satellite in the world, to develop algorithms and models of 4-dimensional (4D) cities. They used point cloud technology to make 4D point clouds of Berlin, Las Vegas, Paris and Washington, D.C. These models of cities eventually can help monitor the built environment and deliver timely signals of potential dangers in infrastructures and buildings, e.g. subterranean subsidence.
Picture (above): Las Vegas in 3D. (Image: TUM & DLR).
Picture (above): A 3d-image of Paris. (Image: TUM & DLR).
The highly precise 4D city models (3D plus the time parameter) can make changes of approximately 1mm per year become visualized.
In TUM, the research team made use of the satellite which flies over the region of interest every 11 days. By applying the same principle as used by computer tomography for producing three-dimensional view of the inside of human body, a variety of radar images taken from different perspectives are combined to create a three-dimensional image.
Experience the Tomographic City Tour of Berlin from Space.
As for the 4th dimension, time, it came with the realization that cities are never static. They are living organisms. Not only because constructions are always in progress, built structures are in motion from time to time, caused by the underground conditions, wind effects and other factors.
Point cloud technology
A point cloud is a set of data points in space. Point clouds are produced by measuring a large number of points on the external surfaces of objects with 3D scanners. They are used for creating 3D models which deliver precise models of reality at reasonable costs. The high-performance processing ability allows the entire environment and individual components, including all built assets, to be captured accurately. The 4D point clouds that combine 3D point clouds capture at different points in time enables a unique spatio-temporal model to be set up.