Deep Convolutional Neural Networks with Transfer Learning for Old Buildings Pathologies Automatic Detection
- 27 Downloads
The automatic detection of structure defects based on computer vision is evolving, especially with constant advances in Deep Convolutional Neural Network. Several image-processing methods have been proposed over the years based on Deep Learning. Nevertheless, the studies are mainly concerned with the crack damage and do not takes into account the other pathologies that can affect a surface structure such as Alkali-silica reaction (ASR), efflorescence, carbonation of concrete, friable plaster, water infiltration, and scaling. In this paper, we propose the method of pre-trained learning Deep Convolutional Neural Networks DCNN model with Transfer learning for the detection of seven classes of old building damage in Medina of Fez and Meknes in Morocco. The robustness of the proposed approach is tested on different architectures and a small set of images not used in learning and validation steps.
KeywordsStructure defects Computer vision Deep learning Transfer learning Artificial intelligence
We thank our partners from Structural Engineering Research and Consulting Office “Structal Engineering Bureau d’études & Conseil”, Fez Agency 134 B Lot Riad Azzaitoun R.A.C, Etg1, Route Ain Chkef 30000 Fez-Morocco, and our partners from NBR-North Center Laboratory of Building and Public Works, 29 Rue Abdessalam Mestaoui, quartier Adarissa, VN Fez- Morocco, who provided insight, expertise and assistance with defects identification and relevance that greatly assisted our research.
This work was supported by CMCLI Project, funded by My Ismaïl University Meknes.
- 1.Fez Old Medina Revamp. Ministry of Communication and Culture. Electronic, Kingdom of Morocco. Sources (Web Publications), 15 April 2019. http://www.maroc.ma/en/royal-activities/fez-old-medina-revamp-unwavering-will-hm-king-preserve-ancient-city. Accessed 31 May 2019
- 5.Faghih-Roohi, S., Hajizadeh, S., Núñez, A., Babuska, R., De Schutter, B.: Deep convolutional neural networks for detection of rail surface defects. In: 2016 International Joint Conference on Neural Networks (IJCNN), pp. 2584–2589. IEEE, July 2016Google Scholar
- 6.Zhang, L., Yang, F., Zhang, Y.D., Zhu, Y.J.: Road crack detection using deep convolutional neural network. In: 2016 IEEE International Conference on Image Processing (ICIP), pp. 3708–3712. IEEE, September 2016Google Scholar
- 10.Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 580–587 (2014)Google Scholar
- 11.Girshick, R.: Fast R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1440–1448 (2015)Google Scholar
- 12.Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp. 91–99 (2015)Google Scholar
- 13.Dai, J., Li, Y., He, K., Sun, J.: R-FCN: object detection via region-based fully convolutional networks. In: Advances in Neural Information Processing Systems, pp. 379–387 (2016)Google Scholar
- 14.Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 779–788 (2016)Google Scholar
- 15.Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., Berg, A.C.: SSD: single shot multibox detector. In: European Conference on Computer Vision, pp. 21–37. Springer, Cham, October 2016Google Scholar
- 16.He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2961–2969 (2017)Google Scholar