Damage evaluation is an important index for the evaluation of buildings health. To provide a rapid crack
evaluation in practical applications, a crack identification and damage evaluation via deep learning
framework is proposed in this paper. We built a combined dataset from Kaggle and site photos. A pre-trained
U-net model is used to perform the training of model. With updated weights, the identification of cracks
could be performed on non-labelled photos.
Cite This Article
APA Style
Xu, S., Tang, H., Wang, D., Zhu, R., Wang, L. et al. (2023). Damage evaluation of building surface via novel deep learning framework. The International Conference on Computational & Experimental Engineering and Sciences, 27(4), 1-3. https://doi.org/10.32604/icces.2023.09930
Vancouver Style
Xu S, Tang H, Wang D, Zhu R, Wang L, Hao S. Damage evaluation of building surface via novel deep learning framework. Int Conf Comput Exp Eng Sciences . 2023;27(4):1-3 https://doi.org/10.32604/icces.2023.09930
IEEE Style
S. Xu, H. Tang, D. Wang, R. Zhu, L. Wang, and S. Hao "Damage Evaluation of Building Surface via Novel Deep Learning Framework," Int. Conf. Comput. Exp. Eng. Sciences , vol. 27, no. 4, pp. 1-3. 2023. https://doi.org/10.32604/icces.2023.09930