EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Coarse-to-Fine Intelligent Inspection Framework for Building Fire Hazard Recognition

Song Ye, Yuting Liu, Chunjin Yu, Jialei Chen, Xili Wan, Lu Wang, Guangming Zhang

Buildings · 15.05.2026 · 16 (10) · 1958

DOI: 10.3390/buildings16101958 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Buildings, online erschienen am 15.05.2026. Verfasst von Song Ye und 6 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Nanjing Urban and Rural Construction Commission.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Building fire safety inspection is a knowledge-intensive engineering task that requires reliable hazard recognition under complex visual conditions, limited labeled data, and strict regulatory accountability. To address these challenges, this paper proposes a coarse-to-fine intelligent inspection framework for building fire hazard recognition and regulation-grounded reporting. The framework first performs binary hazard screening and then refines positive or uncertain cases into specific hazard categories, thereby aligning the inference process with practical inspection workflows. A self-supervised DINOv2 Vision Transformer is adopted as the visual backbone, and a small-sample adaptation strategy is developed by combining staged fine-tuning, a lightweight SE-based classification head, and task-aligned knowledge distillation. In addition, an Agentic RAG compliance layer is introduced to retrieve, verify, and present clause-level regulatory evidence while suppressing hallucinated or unverifiable citations. Experiments on a real-world building fire hazard image dataset show that the proposed framework achieves stable recognition performance, outperforms representative CNN-, supervised Transformer-, and self-supervised Transformer-based baselines, and improves the faithfulness of regulation-grounded reporting. The results suggest that the proposed framework provides a feasible prototype-level pathway toward intelligent and auditable fire safety inspection, while broader multi-site validation and robustness evaluation remain necessary for future deployment.

Förderung

  • Nanjing Urban and Rural Construction Commission

Zitieren

APA

Ye, S., Liu, Y., Yu, C., Chen, J., Wan, X., Wang, L., & Zhang, G. (2026). A Coarse-to-Fine Intelligent Inspection Framework for Building Fire Hazard Recognition. Buildings, 16(10), 1958. https://doi.org/10.3390/buildings16101958

Vancouver

Ye S, Liu Y, Yu C, Chen J, Wan X, Wang L, et al. A Coarse-to-Fine Intelligent Inspection Framework for Building Fire Hazard Recognition. Buildings. 2026;16(10):1958. doi:10.3390/buildings16101958

BibTeX
@article{Ye2026,
  author = {Ye, Song and Liu, Yuting and Yu, Chunjin and Chen, Jialei and Wan, Xili and Wang, Lu and Zhang, Guangming},
  title = {A Coarse-to-Fine Intelligent Inspection Framework for Building Fire Hazard Recognition},
  journal = {Buildings},
  year = {2026},
  volume = {16},
  number = {10},
  pages = {1958},
  doi = {10.3390/buildings16101958}
}
Kostenlos starten →