The impact of using multi-source remote sensing images on building segmentation with U-Net model

• Autorzy: Pham Trung Dung , Pham Ngoc Hung

Identyfikatory

DOI

10.29227/IM-2025-01-02-070

• Konferencja: CESD 2024 : Conference on Earth Sciences : November 11th, 2024, Ho Chi Minh City, Vietnam
• Języki publikacji: EN

Abstrakty

EN

The building extraction from remote sensing (RS) images has been a significant area of research in the photogrammetric and remote sensing communities, especially with the development of deep learning for over a decade. With the availability of multi-source data from RS images, accurately identifying buildings with different spatial image resolutions has become a challenging task. In this study, we assessed how the unalignment of image resolution between the training and testing datasets affects the ability to extract buildings. Image resolution plays a crucial role in the performance of building extraction. Our experiments found that as the image resolution decreased from 10 cm to 50 cm, the efficiency of building segmentation reduced from 0.759 to 0.585 according to the IoU metric. Besides, the ability and accuracy of building segmentation significantly decreased when the difference in image resolution between the training and testing datasets increased. In the case study, we use the model trained on a 10 cm resolution dataset to predict for 50 cm resolution data, the IoU drops significantly to 0.299. This research offers important insights into building segmentation tasks using multi-source data from satellite, airborne, and UAV images.

Słowa kluczowe

EN

semantic segmentation; building extraction; U-Net model; multi-source remote sensing images; Image resolution

PL

segmentacja semantyczna; ekstrakcja budynków; model U-Net; obrazy teledetekcyjne z wielu źródeł; rozdzielczość obrazu

• Wydawca: Polskie Towarzystwo Przeróbki Kopalin
• Czasopismo: Inżynieria Mineralna
• Rocznik: 2025
• Tom: Vol. 2, no. 1
• Strony: art. no. 070
• Opis fizyczny: Bibliogr. 24 poz., rys., tab., wykr., zdj.

Twórcy

autor

Pham Trung Dung

• Hanoi University of Mining and Geology, Hanoi, Vietnam

autor

Pham Ngoc Hung

• Phenikaa University, Hanoi, Vietnam

Bibliografia

• 1. Badrinarayanan, V., A. Handa and R. J. a. p. a. Cipolla (2015). "Segnet: A deep convolutional encoderdecoder architecture for robust semantic pixel-wise labelling."
• 2. Boonpook, W., Y. Tan and B. J. I. J. o. R. S. Xu (2021). "Deep learning-based multi-feature semantic segmentation in building extraction from images of UAV photogrammetry." 42(1): 1-19.
• 3. Guo, Z., X. Shi, H. Zhang, D. Huang, X. Song, J. Yan, and R. J. a. p. a. Shibasaki (2023). "Enhancing building semantic segmentation accuracy with super resolution and deep learning: Investigating the impact of spatial resolution on various datasets."
• 4. Guo, Z., G. Wu, X. Song, W. Yuan, Q. Chen, H. Zhang, X. Shi, M. Xu, Y. Xu, and R. J. I. A. Shibasaki (2019). "Super-resolution integrated building semantic segmentation for multi-source remote sensing imagery." 7: 99381-99397.
• 5. LeCun, Y., L. Bottou, Y. Bengio and P. J. P. o. t. I. Haffner (1998). "Gradient-based learning applied to document recognition." 86(11): 2278-2324.
• 6. Lee, D. H., K. M. Lee, S. U. J. P. E. Lee, and R. Sensing (2008). "Fusion of lidar and imagery for reliable building extraction." 74(2): 215-225.
• 7. Li, J., X. Huang, L. Tu, T. Zhang, L. J. G. Wang and R. Sensing (2022). "A review of building detection from very high resolution optical remote sensing images." 59(1): 1199-1225.
• 8. Li, Z., Q. Xin, Y. Sun and M. J. R. S. Cao (2021). "A deep learning-based framework for automated extraction of building footprint polygons from very high-resolution aerial imagery." 13(18): 3630.
• 9. Liu, M., T. Yu, X. Gu, Z. Sun, J. Yang, Z. Zhang, X. Mi, W. Cao and J. J. R. S. Li (2020). "The impact of spatial resolution on the classification of vegetation types in highly fragmented planting areas based on unmanned aerial vehicle hyperspectral images." 12(1): 146.
• 10. Long, J., E. Shelhamer and T. Darrell (2015). Fully convolutional networks for semantic segmentation. Proceedings of the IEEE conference on computer vision and pattern recognition.
• 11. Lv, B., L. Peng, T. Wu and R. Chen (2020). Research on urban building extraction method based on deep learning convolutional neural network. IOP Conference Series: Earth and Environmental Science, IOP Publishing.
• 12. Mo, J. S., S. K. Seong, J. W. J. J. o. t. K. S. o. S. Choi, Geodesy, Photogrammetry and Cartography (2021). "Comparative evaluation of deep learning-based building extraction techniques using aerial images." 39(3): 157-165.
• 13. P, P. S., J. Soni and H. A. B (2022). Building extraction from remote sensing images using deep learning and transfer learning. IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium.
• 14. Pilinja Subrahmanya, P., B. Haridas Aithal and S. J. J. o. t. I. S. o. R. S. Mitra (2021). "Automatic extraction of buildings from uav-based imagery using artificial neural networks." 49(3): 681-687.
• 15. Raghavan, R., D. C. Verma, D. Pandey, R. Anand, B. K. Pandey and H. Singh (2022). "Optimized building extraction from high-resolution satellite imagery using deep learning." Multimedia Tools and Applications 81(29): 42309-42323.
• 16. Raghavan, R., D. C. Verma, D. Pandey, R. Anand, B. K. Pandey, H. J. M. T. Singh and Applications (2022). "Optimized building extraction from high-resolution satellite imagery using deep learning." 81(29): 42309-42323.
• 17. Ronneberger, O., P. Fischer and T. Brox (2015). U-net: Convolutional networks for biomedical image segmentation. Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18, Springer.
• 18. Roth, K. L., D. A. Roberts, P. E. Dennison, S. H. Peterson and M. J. R. s. o. e. Alonzo (2015). "The impact of spatial resolution on the classification of plant species and functional types within imaging spectrometer data." 171: 45-57.
• 19. Stiller, D., T. Stark, M. Wurm, S. Dech and H. Taubenböck (2019). Large-scale building extraction in very high-resolution aerial imagery using mask r-cnn. 2019 Joint Urban Remote Sensing Event (JURSE), IEEE.
• 20. Tejeswari, B., S. K. Sharma, M. Kumar, K. J. T. I. A. o. t. P. Gupta, Remote Sensing and S. I. Sciences (2022). "Building footprint extraction from space-borne imagery using deep neural networks." 43: 641-647.
• 21. Wu, Y. (2022). Deep learning based building extraction from high-resolution remote sensing images, University of Waterloo.
• 22. Yu, D., S. Ji, J. Liu, S. J. I. J. o. P. Wei and R. Sensing (2021). "Automatic 3d building reconstruction from multi-view aerial images with deep learning." 171: 155-170.
• 23. Zhang, P., H. He, Y. Wang, Y. Liu, H. Lin, L. Guo and W. J. I. A. Yang (2022). "3d urban buildings extraction based on airborne lidar and photogrammetric point cloud fusion according to u-net deep learning model segmentation." 10: 20889-20897.
• 24. Zheng, L., P. Ai and Y. Wu (2020). Building recognition of uav remote sensing images by deep learning. IGARSS 2020-2020 IEEE International Geoscience and Remote Sensing Symposium, IEEE.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki i promocja sportu (2026).