Kermani, Mohammad Mehdi (2025) Automated Window-to-Wall Ratio Estimation Using Google 3D Map Tiles Analysis and AI. Masters thesis, Concordia University - Next Generation Cities Institue.
![[thumbnail of Kermani_MA_S2025.pdf]](https://spectrum.library.concordia.ca/style/images/fileicons/text.png) |
Text (application/pdf)
4MBKermani_MA_S2025.pdf - Accepted Version Restricted to Repository staff only until 17 April 2027. Available under License Spectrum Terms of Access. |
Abstract
The Window-to-Wall Ratio (WWR) is a critical parameter in architecture and urban planning, influencing energy efficiency, daylighting, and thermal performance. Traditionally, WWR is either manually measured or obtained from architectural plans, but these methods are impractical for large-scale studies due to data unavailability. Recent advances in machine learning and 3D mapping offer new opportunities for automation. In this work, we propose an automated method to estimate WWR by extracting 3D building meshes from Google 3D Maps and using a machine learning model to segment windows and walls from their images. We utilize Google 3D Maps for 3D building model extraction and images, OpenStreetMap (OSM) for defining building boundaries, and a UNet-based deep learning model to segment windows, walls, roofs, and other elements. The UNet model, trained on 1,000 manually labeled building images, accurately segments windows and walls, from which WWR is calculated. Unlike previous methods relying on facade images, drone surveys, or street-level views, our approach leverages Google’s pre-existing 3D city data, available for over 2,500 cities worldwide, offering a scalable, effective, and cost-efficient solution. The AI-predicted WWR values are validated by comparing them with manual measurements from actual buildings, confirming the method’s accuracy. This research advances the automation of architectural and energy analysis, providing a fast, cost-effective, machine learning-driven tool for WWR estimation. Future work includes enhancing model robustness, incorporating larger datasets, and exploring additional 3D data sources.
| Divisions: | Concordia University > Gina Cody School of Engineering and Computer Science > Concordia Institute for Information Systems Engineering |
|---|---|
| Item Type: | Thesis (Masters) |
| Authors: | Kermani, Mohammad Mehdi |
| Institution: | Concordia University - Next Generation Cities Institue |
| Degree Name: | M.A. Sc. |
| Program: | Quality Systems Engineering |
| Date: | 24 March 2025 |
| Thesis Supervisor(s): | Eicker, Ursula |
| Keywords: | Window-to-Wall Ratio, 3D Building Mesh, Machine Learning, Google 3D Map Tiles, UNet, AI |
| ID Code: | 995293 |
| Deposited By: | Mohammad Mehdi Kermani |
| Deposited On: | 17 Jun 2025 17:17 |
| Last Modified: | 17 Jun 2025 17:17 |
References:
Azari, R., & Pena, B. (2019). Window-to-wall ratio and energy use: A review on performance-based design approaches. Energy and Buildings, 203, 109399. https://doi.org/10.1016/j.enbuild.2019.109399Barron, C., Neis, P., & Zipf, A. (2014). A comprehensive framework for assessing the quality of OSM building data. International Journal of Geographical Information Science, 28(9), 1–19. https://doi.org/10.1080/13658816.2014.897174
Biljecki, F., Ledoux, H., & van Oosterom, P. (2017). The importance of height data for 3D city models: A case study in the Netherlands. Computers, Environment and Urban Systems, 64, 1–17. https://doi.org/10.1016/j.compenvurbsys.2017.02.005
Blender Foundation. (2024). Blender (Version 4.3.2) [Computer software]. https://www.blender.org/
Blosm Plugin. (n.d.). Blender OpenStreetMap & Google 3D Tiles Plugin Documentation. Retrieved from https://blender-osm.com/
Dai, F., Aksamija, A., & Hong, T. (2021). Automated facade segmentation using deep learning for large-scale building analysis. Building and Environment, 199, 107921. https://doi.org/10.1016/j.buildenv.2021.107921
De Simone, M., Al-Hussein, M., & Sattar, S. (2023). Estimating window-to-wall ratio using deep learning on Google Street View imagery. Applied Energy, 342, 120811. https://doi.org/10.1016/j.apenergy.2023.120811
Fan, H., Zipf, A., Fu, Q., & Neis, P. (2014). Quality assessment of OSM building footprints in Munich. International Journal of Digital Earth, 7(9), 1–16. https://doi.org/10.1080/17538947.2013.847125
Google. (n.d.). Google Maps Platform Terms of Service. Retrieved from https://cloud.google.com/maps-platform/terms
Google Developers. (n.d.). Google Maps Platform documentation. Retrieved from https://developers.google.com/maps
Goetz, M. (2013). Towards generating highly detailed 3D CityGML models from OpenStreetMap. ISPRS International Journal of Geo-Information, 2(4), 603–625. https://doi.org/10.3390/ijgi2040603
Gu, D., & Choo, H. (2022). Method for constructing a facade dataset through deep learning-based automatic image labeling. Automation in Construction, 135, 104112. https://doi.org/10.1016/j.autcon.2022.104112
Kadhim, N., & Mourshed, M. (2016). Estimating building heights from Google Earth images. Energy & Buildings, 118, 86–95. https://doi.org/10.1016/j.enbuild.2016.02.048
Kämpf, J. H., & Robinson, D. (2010). A simplified approach to urban energy modeling. Energy and Buildings, 42(10), 2058–2065. https://doi.org/10.1016/j.enbuild.2010.06.016
Kim, Y., & Boyer, L. (2017). Impact of window-to-wall ratio on energy consumption and daylighting in office buildings: A case study. Sustainability, 9(8), 1281. https://doi.org/10.3390/su9081281
Laumer, M., Musialski, P., & Wimmer, M. (2019). Towards automatic urban facade analysis using Google Street View. Computers, Environment and Urban Systems, 77, 101350. https://doi.org/10.1016/j.compenvurbsys.2019.101350
OpenStreetMap Contributors. (n.d.). OpenStreetMap data. Retrieved from https://www.openstreetmap.org
Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 234–241). Springer. https://arxiv.org/abs/1505.04597
Schiefelbein, J., Peters, R., & Müller, A. (2019). Automatic 3D building reconstruction using OSM data for urban energy simulations. Energy and Buildings, 183, 243–256. https://doi.org/10.1016/j.enbuild.2019.01.039
Touzani, S., & Fernandes, C. (2021). A machine learning approach to estimate window-to-wall ratio using drone imagery. Automation in Construction, 126, 103677. https://doi.org/10.1016/j.autcon.2021.103677
Tzempelikos, A., & Athienitis, A. K. (2007). The impact of shading design and control on building cooling and lighting demand. Solar Energy, 81(3), 369-382. https://doi.org/10.1016/j.solener.2006.06.015
Yao, J., Zhao, J., & Li, X. (2019). Optimum window-to-wall ratio for high-rise residential buildings in hot summer and cold winter zones of China. Journal of Building Performance Simulation, 12(1), 1-14. https://doi.org/10.1080/19401493.2018.1540896
Zhao, X., Li, J., & Zhang, H. (2023). Google Earth and LiDAR integration for estimating building heights at city scale. International Journal of Applied Earth Observation and Geoinformation, 117, 103213. https://doi.org/10.1016/j.jag.2023.103213
Zhou, X., Biljecki, F., & Ledoux, H. (2023). Global completeness of OSM building data: A comparative analysis. International Journal of Digital Earth, 16(1), 1–15. https://doi.org/10.1080/17538947.2022.2153174
Zhuo, W., Zhang, Y., & Kopp, R. (2019). Building facade segmentation from oblique UAV imagery using deep learning. ISPRS Journal of Photogrammetry and Remote Sensing, 149, 159-169. https://doi.org/10.1016/j.isprsjprs.2019.01.004
Zhuo, W., Zhang, Y., & Kopp, R. (2023). Direct window-to-wall ratio prediction using deep learning approaches. Remote Sensing, 15(2), 452. https://doi.org/10.3390/rs15020452
Repository Staff Only: item control page
Download Statistics
Download Statistics