EUVIMEDEUVIMED entdecken
Crossref

CNN-LSTM detection and short-term prediction of concrete crack severity

Jun Tang, Guolin Guo, Ting Wang, Qi Wang, Qingyi Meng, Shuxin Ni

Proceedings of the Institution of Civil Engineers - Construction Materials · 04.11.2026 · 179 (6) · 473-484

DOI: 10.1680/jcoma.25.00109 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the Institution of Civil Engineers - Construction Materials, online erschienen am 20.08.2026. Verfasst von Jun Tang und 5 weiteren Autor:innen. Die Arbeit nennt 24 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The development of concrete cracks will affect the durability, serviceability, and safety of civil infrastructure. This study proposes a leakage-controlled convolutional neural network long/short-term memory (LSTM) framework for static crack detection and short-term prediction of crack severity from time-ordered image sequences. The temporal task is defined as a next-step prediction of the crack severity index (CSI) derived from the image, rather than claiming complete physical crack geometry prediction from bounding box annotations. The CSI is calculated based on the ratio of crack-box and region-of-interest areas. The ResNet34 encoder extracts frame-wise spatial features, and a two-layer LSTM models the temporal dependency of structure-level sequences. Sequence construction, augmentation usage, split rules, optimisation settings, and evaluation metrics are selected to improve reproducibility and avoid temporal leakage. On the dataset, the framework achieves 93.0% classification accuracy, 88.3% mean average precision (mAP)@0.5, 0.76 mean intersection over union, 0.015 normalised mean squared error, and 89.2% F1 score. It also clarifies the practical interpretation of CSI growth, discusses detector-level benchmarking, domain shift and uncertainty estimation, and takes this method as an interpretable short-horizon monitoring framework, rather than a fully calibrated crack metrology system.

Zitieren

APA

Tang, J., Guo, G., Wang, T., Wang, Q., Meng, Q., & Ni, S. (2026). CNN-LSTM detection and short-term prediction of concrete crack severity. Proceedings of the Institution of Civil Engineers - Construction Materials, 179(6), 473-484. https://doi.org/10.1680/jcoma.25.00109

Vancouver

Tang J, Guo G, Wang T, Wang Q, Meng Q, Ni S. CNN-LSTM detection and short-term prediction of concrete crack severity. Proceedings of the Institution of Civil Engineers - Construction Materials. 2026;179(6):473-484. doi:10.1680/jcoma.25.00109

BibTeX
@article{Tang2026,
  author = {Tang, Jun and Guo, Guolin and Wang, Ting and Wang, Qi and Meng, Qingyi and Ni, Shuxin},
  title = {CNN-LSTM detection and short-term prediction of concrete crack severity},
  journal = {Proceedings of the Institution of Civil Engineers - Construction Materials},
  year = {2026},
  volume = {179},
  number = {6},
  pages = {473--484},
  doi = {10.1680/jcoma.25.00109}
}
Kostenlos starten →