Deep Learning with Semantic Segmentation Approach for Building Rooftop Mapping in Urban Irregular Housing Complexes
Received: 20 November 2024 | Revised: 26 December 2024 | Accepted: 29 December 2024 | Online: 3 April 2025
Corresponding author: Edy Irwansyah
Abstract
This research investigates the application of the Deep Learning (DL) U-Net architecture for building rooftop segmentation in densely populated urban areas with irregular housing patterns. The research explores the effectiveness of two loss functions - Binary Cross Entropy (BCE) and Dice Loss (DLs) - to optimize the segmentation accuracy. The present study utilized Small-Format Aerial Photography (SFAP) images processed into orthophotos with a final ground sampling distance of 5 cm. The study area, located in Bogor, Indonesia, features both regular and irregular housing patterns, making it an ideal testing ground for the segmentation model. The U-Net model, having been utilized EfficientNetB6 as the encoder and having been trained with augmented data, demonstrated stable performance across metrics, such as accuracy, precision, recall, and F1-score. The results show that the DLs function outperformed BCE, achieving an average Intersection over Union (IoU) score of 96.8% compared to the 87% score for BCE, indicating that DLs is more effective for this application. The study further enhances the segmentation results by converting the raster data into a vector format using the Ramer-Douglas-Peucker (RDP) algorithm, which simplifies and smooths the polygonal shapes of the segmented rooftops. The combination of the U-Net, DLs and RDP algorithm provides high accuracy results and high usability of the segmentation outputs in practical applications, such as urban planning and disaster management scenarios where accurate rooftop delineation is critical.
Keywords:
deep learning, building rooftop, semantic segmentation, drone data, urban mappingDownloads
References
H. He, L. Ma, and J. Li, "HigherNet-DST: Higher-Resolution Network With Dynamic Scale Training for Rooftop Delineation," IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–15, Feb. 2024.
K. Sawa, I. Yalcin, and S. Kocaman, "Building Detection from SkySat Images with Transfer Learning: a Case Study over Ankara," PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, vol. 92, no. 2, pp. 163–175, Apr. 2024.
L. I. U. Wentao, L. I. Shihua, and Q. I. N. Yuchu, "Automatic Building Roof Extraction with Fully Convolutional Neural Network," Journal of Geo-information Science, vol. 20, no. 11, pp. 1562–1570, Nov. 2018.
H. He et al., "The Impact of Data Volume on Performance of Deep Learning Based Building Rooftop Extraction Using Very High Spatial Resolution Aerial Images," in 2021 IEEE International Geoscience and Remote Sensing Symposium, Brussels, Belgium, 2021, pp. 1343–1346.
J. Yang, B. Matsushita, and H. Zhang, "Improving building rooftop segmentation accuracy through the optimization of UNet basic elements and image foreground-background balance," ISPRS Journal of Photogrammetry and Remote Sensing, vol. 201, pp. 123–137, Jul. 2023.
I. García-Aguilar, J. Galeano-Brajones, F. Luna-Valero, J. Carmona-Murillo, J. D. Fernández-Rodríguez, and R. M. Luque-Baena, "Prediction of Optimal Locations for 5G Base Stations in Urban Environments Using Neural Networks and Satellite Image Analysis," in Bioinspired Systems for Translational Applications: From Robotics to Social Engineering: 10th International Work-Conference on the Interplay Between Natural and Artificial Computation,Proceedings, Part II, Olhâo, Portugal, 2024, pp. 33–43.
G. Li et al., "A district-scale spatial distribution evaluation method of rooftop solar energy potential based on deep learning," Solar Energy, vol. 268, Jan. 2024, Art. no. 112282.
M. Cermak and V. Skala, "Edge spinning algorithm for implicit surfaces," Applied Numerical Mathematics, vol. 49, no. 3–4, pp. 331–342, Jun. 2004.
M. Cermak and V. Skala, "Surface Curvature Estimation for Edge Spinning Algorithm," in Computational Science - ICCS 2004: Proceedings of 4th International Conference on Computational Science, Part II, Kraków, Poland, 2004, pp. 412–418.
D. Harbinson, R. Balsys, and K. Suffern, "Hybrid Polygon-Point Rendering of Singular and Non-Manifold Implicit Surfaces," in 2019 23rd International Conference in Information Visualization – Part II, Adelaide, Australia, 2019, pp. 160–166.
C. Campoverde, M. Koeva, C. Persello, K. Maslov, W. Jiao, and D. Petrova-Antonova, "Automatic Building Roof Plane Extraction in Urban Environments for 3D City Modelling Using Remote Sensing Data," Remote Sensing, vol. 16, no. 8, Apr. 2024, Art. no. 1386.
A. Ramalingam, V. Srivastava, S. V. George, S. Alagala, and M. L. Manickam, "Building rooftop extraction from aerial imagery using low complexity UNet variant models," Journal of Spatial Science, vol. 69, no. 3, pp. 773–800, Jul. 2024.
M. D. Hossain and D. Chen, "Performance Comparison of Deep Learning (DL)-Based Tabular Models for Building Mapping Using High-Resolution Red, Green, and Blue Imagery and the Geographic Object-Based Image Analysis Framework," Remote Sensing, vol. 16, no. 5, Mar. 2024, Art. no. 878.
H. Ni et al., "Enhancing rooftop solar energy potential evaluation in high-density cities: A Deep Learning and GIS based approach," Energy and Buildings, vol. 309, Apr. 2024, Art. no. 113743.
X. Han, J. Wang, X. Liu, J. Du, X. Bai, and R. Ji, "PromptNet: Prompt Learning for Roof Photovoltaic Potential Assessment," Journal of Physics: Conference Series, vol. 2755, no. 1, May 2024, Art. no. 012042.
J. Yang, B. Matsushita, and H. Zhang, "Improving building rooftop segmentation accuracy through the optimization of UNet basic elements and image foreground-background balance," ISPRS Journal of Photogrammetry and Remote Sensing, vol. 201, pp. 123–137, Jul. 2023.
A. Churi and D. B. Megherbi, "Methods for Predictive Performance Improvement of Deep Learning Systems for Aerial Building Roof Detection with Multispectral Images," in 2023 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications, Gammarth, Tunisia, 2023, pp. 1–6.
Z. Liu, H. Tang, L. Feng, and S. Lyu, "China Building Rooftop Area: the first multi-annual (2016–2021) and high-resolution (2.5 m) building rooftop area dataset in China derived with super-resolution segmentation from Sentinel-2 imagery," Earth System Science Data, vol. 15, no. 8, pp. 3547–3572, Aug. 2023.
Z. K. Hussain, J. Congshir, Y. X. Xin, and M. R. e Mustafa, "A Comparative Study of PP-LiteSeg, Dual Attention Network, DeeplabV3p and Asymmetric Neural Network for Rooftop Detection in UAV Images." Preprints, Apr. 10, 2024.
M. Buyukdemircioglu, R. Can, and S. Kocaman, "Deep Learning Based Roof Type Classification using Very High Resolution Aerial Imagery," The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLIII-B3-2021, pp. 55–60, Jun. 2021.
P. Chaweewat, "Solar photovoltaic rooftop detection using satellite imagery and deep learning," in 2023 IEEE PES 15th Asia-Pacific Power and Energy Engineering Conference, Chiang Mai, Thailand, 2023, pp. 1–5.
X. Wang, J. Zhang, and L. You, "A Douglas-Peucker Algorithm Combining Node Importance and Radial Distance Constraints," in 2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture, Manchester, United Kingdom, 2021, pp. 265–269.
Jinsong M. A., Jie S., and Shoucheng X. U., "A Parallel Implementation of Douglas-Peucker Algorithm for Real-Time Map Generalization of Polyline Features on Multi-core Processor Computers," Geomatics and Information Science of Wuhan University, vol. 36, no. 12, pp. 1423–1426, Dec. 2011.
J. Ma, S. Xu, Y. Pu, and G. Chen, "A real-time parallel implementation of Douglas-Peucker polyline simplification algorithm on shared memory multi-core processor computers," in 2010 International Conference on Computer Application and System Modeling, Taiyuan, China, 2010, pp. V4-647-V4-652.
X. Song, C. Cheng, C. Zhou, and D. Zhu, "Gestalt-Based Douglas-Peucker Algorithm to Keep Shape Similarity and Area Consistency of Polygons," Sensor Letters, vol. 11, no. 6–7, pp. 1015–1021, Jun. 2013.
Z. Xie, H. Wang, and L. Wu, "The improved Douglas-Peucker algorithm based on the contour character," in 2011 19th International Conference on Geoinformatics, Shanghai, China, 2011, pp. 1–5.
D. H. Douglas and T. K. Peucker, "Algorithms for the reduction of the number of points required to represent a digitized line or its caricature," Cartographica, vol. 10, no. 2, pp. 112–122, Dec. 1973.
D. Patil and S. Jadhav, "Road Segmentation in High-Resolution Images Using Deep Residual Networks," Engineering, Technology & Applied Science Research, vol. 12, no. 6, pp. 9654–9660, Dec. 2022.
Downloads
How to Cite
License
Copyright (c) 2025 Edy Irwansyah, Alexander A. S. Gunawan, Hady Pranoto, Fabian Surya Pramudya, Lucky Fakhriadi
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
