Calculating Shadows with U-Nets for Urban Environments (Short Paper)

Authors Dominik Rothschedl, Franz Welscher , Franziska Hübl , Ivan Majic , Daniele Giannandrea, Matthias Wastian, Johannes Scholz , Niki Popper

Thumbnail PDF


  • Filesize: 0.68 MB
  • 6 pages

Document Identifiers

Author Details

Dominik Rothschedl
  • dwh GmbH, Vienna, Austria
Franz Welscher
  • Insitute of Geodesy, Graz University of Technology, Austria
Franziska Hübl
  • Insitute of Geodesy, Graz University of Technology, Austria
Ivan Majic
  • Insitute of Geodesy, Graz University of Technology, Austria
Daniele Giannandrea
  • dwh GmbH, Vienna, Austria
  • Institute of Information Systems Engineering, TU Vienna, Austria
Matthias Wastian
  • dwh GmbH, Vienna, Austria
Johannes Scholz
  • Insitute of Geodesy, Graz University of Technology, Austria
Niki Popper
  • dwh GmbH, Vienna, Austria
  • Institute of Information Systems Engineering, TU Vienna, Austria

Cite AsGet BibTex

Dominik Rothschedl, Franz Welscher, Franziska Hübl, Ivan Majic, Daniele Giannandrea, Matthias Wastian, Johannes Scholz, and Niki Popper. Calculating Shadows with U-Nets for Urban Environments (Short Paper). In 12th International Conference on Geographic Information Science (GIScience 2023). Leibniz International Proceedings in Informatics (LIPIcs), Volume 277, pp. 63:1-63:6, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)


Shadow calculation is an important prerequisite for many urban and environmental analyses such as the assessment of solar energy potential. We propose a neural net approach that can be trained with 3D geographical information and predict the presence and depth of shadows. We adapt a U-Net algorithm traditionally used in biomedical image segmentation and train it on sections of Styria, Austria. Our two-step approach first predicts binary existence of shadows and then estimates the depth of shadows as well. Our results on the case study of Styria, Austria show that the proposed approach can predict in both models shadows with over 80% accuracy which is satisfactory for real-world applications, but still leaves room for improvement.

Subject Classification

ACM Subject Classification
  • Computing methodologies → Neural networks
  • Neural Net
  • U-Net
  • Residual Net
  • Shadow Calculation


  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    PDF Downloads


  1. Athanasios Angelis-Dimakis, Markus Biberacher, Javier Dominguez, Giulia Fiorese, Sabine Gadocha, Edgard Gnansounou, Giorgio Guariso, Avraam Kartalidis, Luis Panichelli, Irene Pinedo, et al. Methods and tools to evaluate the availability of renewable energy sources. Renewable and sustainable energy reviews, 15(2):1182-1200, 2011. Google Scholar
  2. Sukriti Bhattacharya, Christian Braun, and Ulrich Leopold. An Efficient 2.5D Shadow Detection Algorithm for Urban Planning and Design Using a Tensor Based Approach. ISPRS International Journal of Geo-Information, 10(9):583, September 2021. URL:
  3. Andrew P Bradley. The use of the area under the roc curve in the evaluation of machine learning algorithms. Pattern recognition, 30(7):1145-1159, 1997. Google Scholar
  4. Zoran Cuckovic. Enhancing terrain cartography with natural shadows, 2019. URL:
  5. Zoran Cuckovic. Terrain shading: a qgis plugin for modelling natural illumination over digital terrain models, 2021. URL:
  6. Luciano da F. Costa. Further generalizations of the jaccard index. CoRR, abs/2110.09619, 2021. URL:
  7. Lucas Fidon, Wenqi Li, Luis C Garcia-Peraza-Herrera, Jinendra Ekanayake, Neil Kitchen, Sébastien Ourselin, and Tom Vercauteren. Generalised wasserstein dice score for imbalanced multi-class segmentation using holistic convolutional networks. In Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: Third International Workshop, BrainLes 2017, Held in Conjunction with MICCAI 2017, Quebec City, QC, Canada, September 14, 2017, Revised Selected Papers 3, pages 64-76. Springer, 2018. Google Scholar
  8. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition, 2015. URL:
  9. Fabian Isensee, Jens Petersen, Andre Klein, David Zimmerer, Paul F. Jaeger, Simon Kohl, Jakob Wasserthal, Gregor Koehler, Tobias Norajitra, Sebastian Wirkert, and Klaus H. Maier-Hein. nnu-net: Self-adapting framework for u-net-based medical image segmentation, 2018. URL:
  10. Dominik Müller, Iñaki Soto-Rey, and Frank Kramer. Towards a guideline for evaluation metrics in medical image segmentation. BMC Research Notes, 15(1):1-8, 2022. Google Scholar
  11. Jesús Polo, Nuria Martín-Chivelet, and Carlos Sanz-Saiz. BIPV Modeling with Artificial Neural Networks: Towards a BIPV Digital Twin. Energies, 15(11), 2022. URL:
  12. P. Redweik, C. Catita, and M. Brito. Solar energy potential on roofs and facades in an urban landscape. Solar Energy, 97:332-341, 2013. URL:
  13. Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation, 2015. URL:
  14. Nahian Siddique, Sidike Paheding, Colin P. Elkin, and Vijay Devabhaktuni. U-net and its variants for medical image segmentation: A review of theory and applications. IEEE Access, 9:82031-82057, 2021. URL:
  15. Abdel Aziz Taha and Allan Hanbury. Metrics for evaluating 3d medical image segmentation: analysis, selection, and tool. BMC medical imaging, 15(1):1-28, 2015. Google Scholar
  16. Sander van der Hoog. Deep learning in (and of) agent-based models: A prospectus, 2017. URL:
  17. Yuan Yin, Vincent Le Guen, Jeremie Dona, Emmanuel de Bezenac, Ibrahim Ayed, Nicolas Thome, and Patrick Gallinari. Augmenting physical models with deep networks for complex dynamics forecasting. Journal of Statistical Mechanics: Theory and Experiment, 2021(12):124012, December 2021. URL: