Kentsel Alanlarda Çatı Yüzey Segmentasyonu ve Fotovoltaik Potansiyel Analizinde Görsel Temel Model ve Kural Tabanlı Yaklaşımlar

Yıl 2025, Cilt: 6 Sayı: 1, 119 - 130, 26.03.2025

Samed Özdemir , Ahmet Yavuzdoğan

https://doi.org/10.48123/rsgis.1606873

https://izlik.org/JA38HN88EA

Öz

Bu çalışma, kentsel alanlarda bina çatılarının fotovoltaik (FV) potansiyelinin hava fotoğrafları ve nokta bulutu verileri üzerinden hızlı ve etkin bir şekilde belirlenmesi için iki yöntem sunulmaktadır. İlk yöntemde, hava fotoğraflarından çatı yüzeyleri ve engellerin tespiti için Segment Anything Model (SAM) ve Contrastive Language Image Pre-Training (CLIP) modelleri kullanılmaktadır. İkinci yöntemde ise Light Detection and Ranging (LiDAR) nokta bulutlarından çatı yüzeylerinin tespitinde Random Sample Consensus (RANSAC) ve Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algoritmaları kullanılmaktadır. Çalışmada önerilen ilk yöntem sayesinde güncel derin öğrenme yaklaşımlarının 2.5B FV potansiyel analizindeki başarımı araştırılırken, önerilen ikinci yaklaşım ile 3B FV potansiyel analizinin 2B yaklaşıma göre başarımı ele alınmaktadır. FV potansiyel analizinde, PhotoVoltaic Geographical Information System (PVGIS) Application Programming Interface (API)’si kullanılmıştır. Önerilen her iki yöntemle elde edilen çatı parametreleri üzerinden analiz edilmektedir. Bina tespitinde, ilk yaklaşım %94.29 IoU skoru elde ederken ikinci yaklaşım ile elde edilen IoU skoru %91.23 olmuştur.

Anahtar Kelimeler

Çatı yüzeyi , Derin öğrenme , Fotovoltaik potansiyel , Nokta bulutu , Semantik segmentasyon

Vision Foundation Models and Rule-Based Approaches for Roof Surface Segmentation and Photovoltaic Potential Analysis in Urban Areas

https://izlik.org/JA38HN88EA

Öz

This study presents two methods for rapidly and effectively determining the photovoltaic (PV) potential of building roofs in urban areas using aerial photographs and point cloud data. In the first method, the Segment Anything Model (SAM) and Contrastive Language Image Pre-Training (CLIP) models are used to detect roof surfaces and obstacles from aerial photographs. In the second method, the Random Sample Consensus (RANSAC) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithms are employed to identify roof surfaces from Light Detection and Ranging (LiDAR) point clouds. Through the first proposed method, the performance of current deep learning approaches in 2.5D PV potential analysis is investigated, while the second approach examines the performance of 3D PV potential analysis compared to the 2D approach. In PV potential analysis, the Photovoltaic Geographical Information System (PVGIS) Application Programming Interface (API) was utilized. The analysis is conducted based on roof parameters obtained through both proposed methods. In building detection, the first approach achieved an Intersection over Union (IoU) score of 94.29%, whereas the second approach attained an IoU score of 91.23%.

Anahtar Kelimeler

Deep learning , Photovoltaic potential , Point cloud , Roof segments , Semantic segmentation

Kaynakça

• Akumu, C. E., & Dennis, S. (2023). Exploring the addition of airborne LiDAR-DEM and derived TPI for urban land cover and land use classification and mapping. Photogrammetric Engineering & Remote Sensing, 89(1), 19–26.
• Ballif, C., Haug, F.-J., Boccard, M., Verlinden, P. J., & Hahn, G. (2022). Status and perspectives of crystalline silicon photovoltaics in research and industry. Nature Reviews Materials, 7(8), 597–616.
• Comert, R., & Kaplan, O. (2018). Object-based building extraction and building period estimation from unmanned aerial vehicle data. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 4(3), 71–76. https://doi.org/10.5194/isprs-annals-IV-3-71-2018
• Cramer, M. (2010). The DGPF-test on digital airborne camera evaluation: Overview and test design. Photogrammetrie-Fernerkundung-Geoinformation, 2, 73–82.
• Ester, M., Kriegel, H.-P., Sander, J., & Xu, X. (1996, August 2–6). A density-based algorithm for discovering clusters in large spatial databases with noise [Conference presentation]. 2nd International Conference on Knowledge Discovery and Data Mining, Portland, Oregon.
• Fischler, M. A., & Bolles, R. C. (1981). Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography. Communications of the ACM, 24(6), 381–395.
• Hansch, R., & Hellwich, O. (2021). Fusion of multispectral LiDAR, hyperspectral, and RGB data for urban land cover classification. IEEE Geoscience and Remote Sensing Letters, 18(2), 366–370.
• Huang, X., Hayashi, K., Matsumoto, T., Tao, L., Huang, Y., & Tomino, Y. (2022). Estimation of rooftop solar power potential by comparing solar radiation data and remote sensing data—A case study in Aichi, Japan. Remote Sensing, 14(7), Article 1742. https://doi.org/10.3390/rs14071742
• Kavzoglu, T., Colkesen, I., Atesoglu, A., Tonbul, H., Yilmaz, E. O., Ozlusoylu, S., & Ozturk, M. Y. (2024). Construction and implementation of a poplar spectral library based on phenological stages for land cover classification using high-resolution satellite images. International Journal of Remote Sensing, 45(6), 2049–2072.
• Kettle, J., Aghaei, M., Ahmad, S., Fairbrother, A., Irvine, S., Jacobsson, J. J., Kazim, S., Kazukauskas, V., Lamb, D., Lobato, K., Mousdis, G. A., Oreski, G., Reinders, A., Schmitz, J., Yilmaz, P., & Theelen, M. J. (2022). Review of technology-specific degradation in crystalline silicon, cadmium telluride, copper indium gallium selenide, dye-sensitised, organic and perovskite solar cells in photovoltaic modules: Understanding how reliability improvements in mature technologies can enhance emerging technologies. Progress in Photovoltaics: Research and Applications, 30(12), 1365–1392. https://doi.org/10.1002/pip.3577
• Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A. C., Lo, W.-Y., Dollár, P., & Girshick, R. (2023). Segment anything. arXiv. http://arxiv.org/abs/2304.02643
• Lan, H., Gou, Z., & Xie, X. (2021). A simplified evaluation method of rooftop solar energy potential based on image semantic segmentation of urban streetscapes. Solar Energy, 230, 912–924.
• Li, S. Y., & Han, J. Y. (2022). The impact of shadow covering on the rooftop solar photovoltaic system for evaluating self-sufficiency rate in the concept of nearly zero energy building. Sustainable Cities and Society, 80, Article 103821. https://doi.org/10.1016/j.scs.2022.103821
• Ma, H., Ma, H., Zhang, L., Liu, K., & Luo, W. (2022). Extracting urban road footprints from airborne LiDAR point clouds with PointNet++ and two-step post-processing. Remote Sensing, 14(3), Article 789. https://doi.org/10.3390/rs14030789
• Mao, H., Chen, X., Luo, Y., Deng, J., Tian, Z., Yu, J., Xiao, Y., & Fan, J. (2023). Advances and prospects on estimating solar photovoltaic installation capacity and potential based on satellite and aerial images. Renewable and Sustainable Energy Reviews, 179, Article 113276. https://doi.org/10.1016/j.rser.2023.113276
• Minelli, F., D’Agostino, D., Migliozzi, M., Minichiello, F., & D’Agostino, P. (2023). PhloVer: a modular and integrated tracking photovoltaic shading device for sustainable large urban spaces—preliminary study and prototyping. Energies, 16(15), Article 5786. https://doi.org/10.3390/en16155786
• Moudrý, V., Beková, A., & Lagner, O. (2019). Evaluation of a high-resolution UAV imagery model for rooftop solar irradiation estimates. Remote Sensing Letters, 10(11), 1077–1085.
• Özdemir, S., Akbulut, Z., Karsli, F., & Acar, H. (2021). Automatic extraction of trees by using multiple return properties of the LiDAR point cloud. International Journal of Engineering and Geosciences, 6(1), 20–26.
• Özdemir, S., Yavuzdoğan, A., Bilgilioğlu, B. B., & Akbulut, Z. (2023). SPAN: An open-source plugin for photovoltaic potential estimation of individual roof segments using point cloud data. Renewable Energy, 216, Article 119022. https://doi.org/10.1016/j.renene.2023.119022
• Ozturk, O., Isik, M. S., Sariturk, B., & Seker, D. Z. (2022). Generation of Istanbul road data set using Google Map API for deep learning-based segmentation. International Journal of Remote Sensing, 43(8), 2793–2812.
• Ozturk, O., Isik, M. S., Kada, M., & Seker, D. Z. (2023). Improving road segmentation by combining satellite images and LiDAR data with a feature-wise fusion strategy. Applied Sciences, 13(10), Article 6161. https://doi.org/10.3390/app13106161
• Psiloglou, B. E., Kambezidis, H. D., Kaskaoutis, D. G., Karagiannis, D., & Polo, J. M. (2020). Comparison between MRM simulations, CAMS and PVGIS databases with measured solar radiation components at the Methoni station, Greece. Renewable Energy, 146, 1372–1391. https://doi.org/10.1016/j.renene.2019.07.064
• Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., & Sutskever, I. (2021). Learning transferable visual models from natural language supervision. arXiv. http://arxiv.org/abs/2103.00020
• Ranalder, L., Busch, H., Hansen, T., Brommer, M., Couture, T., Gibb, D., ... & Sverrisson, F. (2021). Renewables in cities: 2021 global status report. REN21 Secretariat. https://www.ren21.net/wp-content/uploads/2019/05/ GSR2021_Full_Report.pdf
• Sahak, A. S., Karsli, F., Gormus, E. T., & Ahmadi, K. (2023). Seasonal monitoring of urban heat island based on the relationship between land surface temperature and land use/cover: A case study of Kabul City, Afghanistan. Earth Science Informatics, 16(1), 845–861.
• Stack, V., & Narine, L. L. (2022). Sustainability at Auburn University: Assessing rooftop solar energy potential for electricity generation with remote sensing and GIS in a Southern US campus. Sustainability, 14(2), Article 626. https://doi.org/10.3390/su14020626
• Suri, M., Huld, T., Cebecauer, T., & Dunlop, E. D. (2008). Geographic aspects of photovoltaics in Europe: Contribution of the PVGIS website. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 1(1), 34–41. https://doi.org/10.1109/JSTARS.2008.2001431
• Tonbul, H., & Kavzoglu, T. (2020). Semi-automatic building extraction from WorldView-2 imagery using Taguchi optimization. Photogrammetric Engineering & Remote Sensing, 86(9), 547–555.
• Topaloğlu, R. H., Aksu, G. A., Ghale, Y. A. G., & Sertel, E. (2022). High-resolution land use and land cover change analysis using GEOBIA and landscape metrics: A case of Istanbul, Turkey. Geocarto International, 37(25), 9071–9097.
• Yagmur, N., Dervisoglu, A., & Bilgilioğlu, B. (2022). Assessment of rapid urbanization effects with remote sensing techniques. In M. Ben Ahmed, A. A. Boudhir, İ. R. Karaș, V. Jain, & S. Mellouli (Eds.), Innovations in Smart Cities Applications (pp. 571–585). https://doi.org/10.1007/978-3-030-94191-8_46
• Zhao, J., He, X., Li, J., Feng, T., Ye, C., & Xiong, L. (2019). Automatic vector-based road structure mapping using multibeam LiDAR. Remote Sensing, 11(14), Article 14. https://doi.org/10.3390/rs11141726
• Zhong, T., Zhang, Z., Chen, M., Zhang, K., Zhou, Z., Zhu, R., Wang, Y., Lü, G., & Yan, J. (2021). A city-scale estimation of rooftop solar photovoltaic potential based on deep learning. Applied Energy, 298, Article 117132. https://doi.org/10.1016/j.apenergy.2021.117132