Falling trees or branches pose a danger to people, but they also cause enormous economic damage. According to Deutsche Bahn, falling trees are already the most common reason for line closures. A new laser-based measurement method combined with AI-supported data evaluation is set to make tree monitoring more efficient and precise in the future. As part of the TreeVitaScan project, the research partners are developing a process that can be used to create a comprehensive data model of the current state of individual trees. Today, municipalities and road operators inspect the tree population every two to three years. During these inspections, experts examine individual trees and take sporadic measurements – a laborious process. Satellite images are also frequently used for assessment.

3D LiDAR data provide information about soil and tree moisture

In addition to weather and environmental data, as well as tree registry data, the moisture levels of individual trees are to be incorporated into the new data model. This moisture data is recorded by a multispectral 3D LiDAR system developed at Fraunhofer IPM from a road or rail vehicle. The scanner data is easier to interpret than camera images – a decisive advantage that also makes real-time applications possible. One of the questions the Fraunhofer researchers are investigating is to what extent the state of vitality can be automatically derived from the LiDAR data. Together with the team from the University of Freiburg, strategies are also being developed to structure the different data streams and to merge them temporally and spatially in order to create a suitable basis for data modeling.