| Project description |
Within the framework of the project, a measurement and
evaluation method is to be developed with which the surface of
a tunnel can be digitized so precisely that deformation
monitoring can be carried out by surface and depth testing. In
addition, all equipment features (e.g. signage, markings, lane
signals, lighting equipment, hydrants, loudspeakers, etc.) will
be automatically detected using deep learning methods. The aim
is to achieve an accuracy of the position and depth measurement
of approx. 1 mm at a speed of up to 65 km/h. For the
development of the planned method, solutions from
photogrammetry, videogrammetry and deep learning will be used.
Through the intelligent combination of overlapping images, 3D
point clouds can be generated and automatically evaluated.
Viscan as a specialist in the field of photogrammetry and HFT
Stuttgart as a specialist in the field of neural networks work
together cooperatively during the development. |
print page