The utilization of underground space is currently influenced by far-reaching transformation processes. In addition to mining, new fields of application are being discussed. These include the construction of underground pumped storage power plants (UPSP) and the final disposal of radioactive waste. For this purpose, three-dimensional acquisition methods can play an important role. Apart from the widely discussed use of LiDAR, photogrammetric methods such as structure from motion (SFM) are also a suitable option with many capabilities. Therefore, an integrated registration and evaluation approach for underground cavities based on structure from motion will be presented. The combination of different existing (open source) software libraries for image alignment and processing of complex three-dimensional models is evaluated in detail. Such an approach could provide profound knowledge of underground space for future exploration and monitoring by simultaneously enabling tasks of surveying, monitoring, and specifically visualization. Powerful processing hardware and high-resolution camera sensors thus also enable acquisition in low-light environments. Image-based visualization methods are particularly suitable for ensuring transparency when conducting large-scale underground projects. Increased transparency can also help to obtain and maintain the social license to operate. This aspect is particularly relevant for highly discussed projects with broad public interest. The presented approach is intended to provide better understanding of ongoing and planned underground activities in general. As a result, the approach can also contribute to the communication of technical issues in the context of current efforts for energy transition.

structure from motion,photogrammetry,subsurface spatial planning,underground monitoring,cavity registration,multi-sensor data fusion