Deprem sonrası uydu görüntülerinden hasarlı binaların ve hasar düzeylerinin belirlenmesine yönelik yapay öğrenme modeli tasarımı: ‘EQDamage4B_Sat’

Öz

Bir deprem ülkesi olan Türkiye’de deprem sonrası kriz yönetiminde, acil müdahale planlamasında vb. süreçlerde geleneksel yöntemlere ek olarak modern yaklaşımların kullanılması kaçınılmazdır. Bu çalışmada modern yaklaşım sınıfına girebilecek olan uçak, insansız hava aracı ve uydu platformlarındaki farklı sensörlerden elde edilen görüntülerden, yenilikçi yapay öğrenme yaklaşımıyla bina bölütleme ve bina hasar sınıflandırma görevlerinin gerçekleştirilmesi konusu ele alınacaktır. Böylece geleneksel yöntemlere göre daha etkin, hızlı ve doğru bir acil müdahale planlaması yapılabilmesine katkı verilmeye çalışılacaktır. Çalışma kapsamında beklenen Büyük Marmara Depremi gibi bundan sonra Türkiye’de olması olası depremler sonrasında uydu görüntülerinden bina hasarlarının belirlenmesine ilişkin nasıl bir yapay öğrenme altyapısı, mimarisi ve modelleri kullanılabileceğine ilişkin kavramsal bir süreç tasarımı yapılmıştır. Farklı yapay öğrenme yaklaşımları, mimarileri ve modelleri bina hasarlarının belirlenmesinde farklı başarımlar sergilemektedir. Çalışmanın amacı da görüntülere dayalı bina bölütlemesi ve hasar düzeyi sınıflandırma görevlerindeki başarımın iyileştirilmesi için yenilikçi bir yapay öğrenme mimari ve model yaklaşımının önerilmesidir.

Anahtar Kelimeler

• Deprem
• Bina hasar tespiti
• Derin öğrenme
• Transformer mimarisi
• Segmentasyon
• Sınıflandırma

Design of an AI model for determining damaged buildings and damage levels from post-earthquake satellite images: ‘EQDamage4B_Sat’

Öz

In Turkey, a country prone to earthquakes, it is inevitable that modern approaches be used in addition to traditional methods in post-earthquake crisis management, emergency response planning, and other processes. This study will address the use of innovative machine learning approaches to perform building segmentation and building damage classification tasks using images obtained from various sensors on aircraft, unmanned aerial vehicles, and satellite platforms. This will contribute to more effective, faster, and more accurate emergency response planning compared to traditional methods. Within the scope of the study, a conceptual process design has been developed regarding the artificial intelligence infrastructure, architecture, and models that can be used to determine building damage from satellite images after earthquakes that are likely to occur in Turkey in the future. Different artificial intelligence approaches, architectures, and models exhibit varying levels of performance in determining building damage. The aim of this study is to propose an innovative artificial intelligence architecture and model approach to improve performance in image-based building segmentation and damage level classification tasks.

Anahtar Kelimeler

• Earthquake
• Building damage assessment
• Deep learning
• Transformer architecture
• Segmentation
• Classification

Kaynakça

1. Badrinarayanan, V., Kendall, A., & Cipolla, R. (2017). Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE transactions on pattern analysis and machine intelligence, 39(12), 2481-2495.
2. Bandara, W. G. C., & Patel, V. M. (2022). A transformer-based siamese network for change detection. IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium, 207-210.
3. Bao, H., Dong, L., Piao, S., & Wei, F. (2021). Beit: Bert pre-training of image transformers. arXiv preprint arXiv:2106.08254.
4. Birkmann, J. (2006). Measuring vulnerability to natural hazards: towards disaster resilient societies. United Nations University Press.
5. Bouchard, I., Rancourt, M. È., Aloise, D., & Kalaitzis, F. (2022). On transfer learning for building damage assessment from satellite imagery in emergency contexts. Remote Sensing, 14(11), 2532.
6. Brunner, D., Lemoine, G., & Bruzzone, L. (2010). Earthquake damage assessment of buildings using VHR optical and SAR imagery. IEEE Transactions on Geoscience and Remote Sensing, 48(5), 2403-2420.
7. Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020). End-to-end object detection with transformers. In European conference on computer vision, 213-229.
8. Chen, L. C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2018). Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40(4), 834-848.

9. Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., Lu, L., Yuille, A. L., & Zhou, Y. (2021). Transunet: Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.04306.
10. Chen, H., Qi, Z., & Shi, Z. (2022). Remote sensing image change detection with transformers. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-14.
11. Cheng, B., Schwing, A., & Kirillov, A. (2021). Per-pixel classification is not all you need for semantic segmentation. Advances in neural information processing systems, 34, 17864-17875.
12. Cheng, B., Misra, I., Schwing, A. G., Kirillov, A., & Girdhar, R. (2022). Masked-attention mask transformer for universal image segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 1290-1299.
13. Dong, L., & Shan, J. (2013). A comprehensive review of earthquake-induced building damage detection with remote sensing techniques. ISPRS Journal of Photogrammetry and Remote Sensing, 84, 85-99.
14. Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterhiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., & Houlsby, N. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929.
15. Dumoulin, V., & Visin, F. (2018). A guide to convolution arithmetic for deep learning. arXiv preprint arXiv:1603.07285.
16. Eğri, A., & Güney, C. (2020). Telekomünikasyon sektöründe kullanılan ek odası kapaklarının sokak düzeyi görüntülerinden tespit edilme başarımının değerlendirilmesi. Jeodezi ve Jeoinformasyon Dergisi, 7(2), 133-154.
17. Everingham, M., Van Gool, L., Williams, C. K., Winn, J., & Zisserman, A. (2010). The pascal visual object classes (voc) challenge. International journal of computer vision, 88(2), 303-338.
18. Fawcett, T. (2006). An introduction to ROC analysis. Pattern recognition letters, 27(8), 861-874.
19. Gong, Z., Zhou, X., & Yao, W. (2024). MultiScale spectral–spatial convolutional transformer for hyperspectral image classification. IET Image Processing, 18(13), 4328-4340.
20. Gonzalez, R. C., & Woods, R. E. (2018). Digital image processing (4th ed.). Pearson: NY.
21. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. The MIT Press.
22. Gu, J., Xie, Z., Zhang, J., & He, X. (2024). Advances in rapid damage identification methods for post-disaster regional buildings based on remote sensing images: A survey. Buildings, 14(4), 898.
23. Gupta, R., Hosfelt, R., Sajeev, S., Patel, N., Goodman, B., Doshi, J., Heim, E., Choset, H., & Gaston, M. (2019). xBD: A dataset for assessing building damage from satellite imagery. arXiv preprint arXiv:1911.09296.
24. He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017). Mask R-CNN. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2), 386-397.
25. He, P., Zhang, D., Chen, W., & Jia, B. (2025). Spatio-temporal distribution estimation and analysis of post-earthquake casualties in urban areas: A case study in Chenghua District, Chengdu. International Journal of Disaster Risk Reduction, 123, 105477.
26. Hu, Y., & Tang, H. (2021). On the generalization ability of a global model for rapid building mapping from heterogeneous satellite images of multiple natural disaster scenarios. Remote Sensing, 13(5), 984.
27. Ibtehaz, N., & Rahman, M. S. (2020). MultiResUNet: Rethinking the U-Net architecture for multimodal biomedical image segmentation. Neural networks, 121, 74-87.
28. Kaiser, P., Wegner, J. D., Lucchi, A., Jaggi, M., Hofmann, T., & Schindler, K. (2017). Learning aerial image segmentation from online maps. IEEE Transactions on Geoscience and Remote Sensing, 55(11), 6054-6068.
29. Kirillov, A., He, K., Girshick, R., Rother, C., & Dollár, P. (2019). Panoptic segmentation. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 9404-9413.
30. Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Ziao, T., Whitehead, S., Berg, A. C., Lo, W., Dollar, P., & Girshick, R. (2023). Segment anything. Proceedings of the IEEE/CVF international conference on computer vision, 4015-4026.
31. Kornblith, S., Shlens, J., & Le, Q. V. (2019). Do better imagenet models transfer better?. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2661-2671.
32. Köroğlu, S. M. (2025). Deprem sonrasında binaların hasar tespitinde kullanılan yapay öğrenme algoritmalarının analizi (Yüksek Lisans Tezi). İstanbul Teknik Üniversitesi, Bilişim Enstitüsü, İstanbul, Türkiye.
33. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84-90.
34. LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324.
35. LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. nature, 521(7553), 436-444.
36. Li, M,, Fu, Y., & Zhang, Y. (2022). Spatial-Spectral Transformer for Hyperspectral Image Denoising. arXiv preprint arXiv:2211.14090.
37. Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2016). Feature Pyramid Networks for Object Detection. arXiv preprint arXiv:1612.03144.
38. Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017). Feature pyramid networks for object detection. Proceedings of the IEEE conference on computer vision and pattern recognition, 2117-2125.
39. Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. Proceedings of the IEEE/CVF international conference on computer vision, 10012-10022.
40. Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, 3431-3440.
41. Nogueira, K., Penatti, O. A., & Dos Santos, J. A. (2017). Towards better exploiting convolutional neural networks for remote sensing scene classification. Pattern Recognition, 61, 539-556.
42. Powers, D. (2011). Evaluation: From Precision, Recall and F-Measure to ROC, Informedness, Markedness & Correlation. Journal of Machine Learning Technologies, 2(1), 37-63.
43. Rahimi, F., Sadeghi-Niaraki, A., Ghodousi, M., & Choi, S. M. (2025). Spatial-temporal modeling of urban resilience and risk to earthquakes. Scientific Reports, 15(1), 8321.
44. Ravi, N., Gabeur, V., Hu, Y.-T., Hu, R., Ryali, C., Ma, T., Khedr, H., Rädle, R., Rolland, C., Gustafson, L., Mintun, E., Pan, J., Vasudev Alwala, K., Carion, N., Wu, C.-Y., Girshick, R., Dollár, P., Feichtenhofer, C., & Fair, M. (2024). Sam 2: Segment anything in images and videos. arXiv preprint arXiv:2408.00714.
45. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention, 234-241. Cham: Springer international publishing.
46. SpaceNet-GitHub. https://github.com/SpaceNetChallenge?spm=a2ty_o01.29997173.0.0.3aa9c921l98Kq7 (Erişim Tarihi: 10 Nisan 2025).
47. Strudel, R., Garcia, R., Laptev, I., & Schmid, C. (2021). Segmenter: Transformer for semantic segmentation. In Proceedings of the IEEE/CVF international conference on computer vision, 7262-7272.
48. Szeliski, R. (2010). Computer Vision. Springer.
49. van Etten, A., Lindenbaum, D., & Bacastow, T. M. (2019). Spacenet: A remote sensing dataset and challenge series. arXiv preprint arXiv:1807.01232.
50. Wang, J., Sun, K., Cheng, T., Jiang, B., Deng, C., Zhao, Y., Liu, D., Mu, Y., Tan, M., Wang, X., Liu, W., & Xiao, B. (2021). Deep high-resolution representation learning for visual recognition. IEEE transactions on pattern analysis and machine intelligence, 43(10), 3349-3364.
51. Xia, H., Wu, J., Yao, J., Zhu, H., Gong, A., Yang, J., Hu, L., & Mo, F. (2023). A deep learning application for building damage assessment using ultra-high-resolution remote sensing imagery in Turkey earthquake. International Journal of Disaster Risk Science, 14(6), 947-962.
52. Xiao, T., Liu, Y., Zhou, B., Jiang, Y., & Sun, J. (2018). Unified perceptual parsing for scene understanding. Proceedings of the European conference on computer vision (ECCV), 418-434.
53. Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J. M., & Luo, P. (2021). SegFormer: Simple and efficient design for semantic segmentation with transformers. Advances in neural information processing systems, 34, 12077-12090.
54. xView-GitHub. https://github.com/DIUx-xView (Erişim Tarihi: 10 Nisan 2025).
55. Yan, H., Zhang, C., & Wu, M. (2022). Lawin transformer: Improving semantic segmentation transformer with multi-scale representations via large window attention. arXiv preprint arXiv:2201.01615.
56. Yang, W., Zhang, X., & Luo, P. (2021). Transferability of convolutional neural network models for identifying damaged buildings due to earthquake. Remote Sensing, 13(3), 504.
57. Yılankıran, F. Ç., & Guney, C. (2021). Emergency response with mobile geosocial sensing in the post‐app era. Transactions in GIS, 25(2), 897-922.
58. Zhang, T., Gao, P., Dong, H., Zhuang, Y., Wang, G., Zhang, W., & Chen, H. (2022). Consecutive pre-training: A knowledge transfer learning strategy with relevant unlabeled data for remote sensing domain. Remote Sensing, 14(22), 5675.
59. Zhao, H., Shi, J., Qi, X., Wang, X., & Jia, J. (2017). Pyramid scene parsing network. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2881-2890.
60. Zheng, S., Lu, J., Zhao, H., Zhu, X., Luo, Z., Wang, Y., Fu, Y., Feng, J., Xiang, T., Torr, P. H. S., & Zhang, L. (2021). Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 6881-6890.
61. Zhu, X. X., Tuia, D., Mou, L., Xia, G. S., Zhang, L., Xu, F., & Fraundorfer, F. (2017). Deep learning in remote sensing: A comprehensive review and list of resources. IEEE geoscience and remote sensing magazine, 5(4), 8-36.
62. URL-1: spacenet.ai – Accelerating Geospatial Machine Learning. https://spacenet.ai/, (Erişim Tarihi: 10 Nisan 2025).
63. URL-2: Güney, C. (2020). Yenilikçi Mekansal Veri Üretim Yöntemleri, https://canerguney.medium.com/yenilik%C3%A7i-mekansal-veri-%C3%BCretim-y%C3%B6ntemleri-1-32c43ff22836, (Erişim Tarihi: 24 Mart 2025).