Evaluation of Building Damage due to Natural Disaster using CNN and GAN

Evaluation of Building Damage due to Natural Disaster using CNN and GAN

Haruka Yamada, Takenori Hida, Xin Wang, Masayuki Nagano

download PDF

Abstract. After destructive natural disasters, it is necessary to quickly grasp the damage situation for the initial response. In recent years, studies on the method of the automatic evaluation of building damages due to disasters using the convolutional neural network (CNN), which is a deep learning methodology for image recognition, were conducted. In these studies, it was clarified that a large number of images are necessary to train the CNN with sufficiently high accuracy. However, the number of images of damaged building is limited. Therefore, in the present study, we used the generative adversarial network (GAN) to automatically generate a large number of imitation images of damaged and undamaged buildings and trained the CNN using imitation images to obtain a higher accuracy rate of the CNN. Then, the validity of the CNN for judgment of “damaged” and “undamaged” using imitation images was confirmed. In addition, photographs of actual buildings were input to the trained CNN as test data.

Building Damage, Deep Learning, Convolutional Neural Networks, Generative Adversarial Networks, Image Recognition, Grad-CAM

Published online 3/30/2023, 9 pages
Copyright © 2023 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA

Citation: Haruka Yamada, Takenori Hida, Xin Wang, Masayuki Nagano, Evaluation of Building Damage due to Natural Disaster using CNN and GAN, Materials Research Proceedings, Vol. 27, pp 67-75, 2023

DOI: https://doi.org/10.21741/9781644902455-9

The article was published as article 9 of the book Structural Health Monitoring

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

[1] Cabinet Office, Government of Japan: Damage from the earthquake centered in Kumamoto region of Kumamoto Prefecture, (as of 8:00 a.m. on April 20) to (as of 5:15 p.m. on June 16), 2016. (in Japanese)
[2] Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner: Gradient-based learning applied to document recognition, Proc. of the IEEE, pp.2278-2324, 1998. https://doi.org/10.1109/5.726791
[3] T. Hida, T. Yaoyama and T. Takada: Building’s Damage Evaluation and its Validation through CNN and Grad-CAM, JCOSSAR 2019, Tokyo, Japan, 2019. (in Japanese)
[4] M. Fujiu, M. Ohara, S. Nakayama, and J. Takayama: Development of Building Damage Assessment Lessening Application Using 3d Modeled House, JSCE Collection of papers (A1) Vol.71, No.4, pp.I_865-I_872, 2015. (in Japanese). https://doi.org/10.2208/jscejseee.71.I_865
[5] M. Yamaguchi, T. Hida, T. Itoi, and T. Takada: Image-Based Building Damage Evaluation Based on Semantic Segmentation and Convolutional Neural Network, The Seventh Asian-Pacific Symposium on Structural Reliability and Its Applications, APSSRA2020, 2020.
[6] H. Yamada, T Hida, X. Wang, T. Yamashita, M. Nagano: Damage evaluation of buildings after natural disaster based on 3DCG and deep learning technique, Summaries of Technical Papers of Annual Meeting AIJ, pp.13-14, 2021. (in Japanese)
[7] I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville and Y. Bengio: Generative adversarial nets, Proc. NIPS 2014, pp.2672–2680, 2014.
[8] B. Liu, Y. Zhu, K. Song, A. Elgammal: Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis, ICLR 2021, 2021.
[9] Y. LeCun, Y. Bengio: Convolutional networks for images, speech, and time series, The Handbook of Brain Theory and Neural Networks, Vol.3361, pp.255-258, 1995.
[10] Y. Jia, E. Shelhamer, J. Donahue, and S. Karayev: Caffe: Convolutional architecture for fast feature embedding, In Proc. ACM Int. Conf. on Multimedia, pp.675-678, 2014. https://doi.org/10.1145/2647868.2654889
[11] A. Krizhevsky, I. Sutskever, G. E. Hinton: ImageNet Classification with Deep Convolutional Neural Networks, In Proc. of NIPS, 2012.
[12] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, Devi Parikh, and D. Batra: Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization, 2017 IEEE International Conference on Computer Vision, 2017. https://doi.org/10.1109/ICCV.2017.74
[13] O. Ronneberger, P. Fischer, and T. Brox: U net: Convolutional networks for biomedical image segmentation, arXiv:1505.04597, 2015. https://doi.org/10.1007/978-3-319-24574-4_28