Use of photogrammetry and terrestrial laser scanning to survey superficial weathering damage to façade building materials - Mainz City Hall

Finished
Test area at the town hall with visualisation of the facade panel deformations in a photograph i3mainz, CC BY SA 4.0

In the past years, we can look back on a successful cooperation with the Institut für Steinkonservierung e.V. (Institute for Stone Conservation). (IfS) can be looked back on. After a number of pilot projects to investigate the usability of current 3D measurement technology for the purpose of monitoring historic stone surfaces were considered in the past, an investigation into the detection of deformed stone surfaces was carried out in 2017 as part of the renovation of Mainz Town Hall.

Motivation

On facades made of natural stone, brick, plaster or concrete, surveying the extent and degree of superficial damage - such as weathering, spalling, bulging or bending - is important with regard to determining restoration measures or assessing traffic safety. Common practice here is regular inspection from a cherry picker or scaffolding. The Mainz Town Hall, a building completed in 1973 according to plans by Arne Jacobsen, has a curtain wall of Norwegian Porsgrunn marble up to 30 m high. Due to the effects of temperature and humidity, many of the marble slabs, which measure approx. 55x75 cm and are only approx. 3 cm thick, show increasing bending with age. This bending puts stress on the fixing anchors and under unfavourable circumstances this leads to the slabs breaking off from the façade. To maintain traffic safety, there have been regular inspections and maintenance work from the cherry picker so far.

Activities

As an alternative to visual inspection using a cherry picker, modern non-contact 3D measurement systems are available. These include Unmanned Aerial Vehicle (UAV) supported photogrammetry and Terrestrial Laser Scanning (TLS). Since both methods have different approaches to generate 3D information, they are subject to different characteristics that need to be considered during the application of the respective method. This is illustrated by means of the façade example through a methodological comparison of both procedures. In the first method, a UAV is used as a sensor platform for a digital photo camera. An overlapping image composite of the object is recorded, which provides the basis for a 3D reconstruction with the evaluation technique “Structure from Motion” (SfM). The result of the evaluation is a referenced 3D point cloud of the façade surface. The recording of the façade by means of TLS was realised by two viewpoints, whereby the result is also a referenced 3D point cloud of the façade surface. For both techniques, the referencing of the 3D data was implemented by independent tacheometric measurements, which are also used to check the measurement data. This results in two measurements with different methods of the same problem, which are available in the identical coordinate system and can therefore be compared with each other. The aim is to detect corresponding deformations of the individual marble slabs, which is possible with both methods.

Results

The quality of the 3D point cloud is much better in the case of the TLS technique because here the 3D coordinates are measured directly on the object surface and not reconstructed as in the alternative SfM method. Among other things, this leads to the fact that the scatter in some areas of the point cloud is very large. The reason for this turned out to be an unfavourable arrangement of the camera points. Global deformations also occur in the SfM method, which can falsify the interpretation of the results. Therefore, when using the SfM technique on such objects and problems in the future, image and flight planning must be carried out based on previously created simulations. This previously defined planning must then be implemented in situ by the UAV, for example with the help of GPS.