EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Evaluating YOLO Models for Efficient Crack Detection in Concrete Structures Using Transfer Learning

Muhammad Sohaib, Muzamal Arif, Jong-Myon Kim

Buildings · 09.12.2024 · 14 (12) · 3928

DOI: 10.3390/buildings14123928 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Buildings, online erschienen am 09.12.2024. Verfasst von Muhammad Sohaib und 2 weiteren Autor:innen. Die Arbeit wurde bisher 48-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ulsan City and Electronics and Telecommunications Research Institute (ETRI), Korea Institute of Energy Technology Evaluation and Planning (KETEP).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The You Only Look Once (YOLO) network is considered highly suitable for real-time object detection tasks due to its characteristics, such as high speed, single-shot detection, global context awareness, scalability, and adaptability to real-world conditions. This work introduces a comprehensive analysis of various YOLO models for detecting cracks in concrete structures, aiming to assist in the selection of an optimal model for future detection and segmentation tasks. The YOLO models are initially trained on a dataset containing both images with and without cracks, producing a generalized model capable of extracting abstract features beneficial for crack detection. Subsequently, transfer learning is employed using a dataset that reflects real-world conditions, such as occlusions, varying crack sizes, and rotations, to further refine the model. Crack detection in concrete remains challenging due to the wide variation in crack sizes, aspect ratios, and complex backgrounds. To achieve optimal performance, we test different versions of YOLO, a state-of-the-art single-shot detector, and aim to balance inference speed and mean average precision (mAP). Our results indicate that YOLOv10 demonstrates superior performance, achieving a mean average precision (mAP) of 74.52% with an inference time of 19.5 milliseconds per image, making it the most effective among the models tested.

Förderung

  • Ulsan City and Electronics and Telecommunications Research Institute (ETRI)
  • Korea Institute of Energy Technology Evaluation and Planning (KETEP)

Zitieren

APA

Sohaib, M., Arif, M., & Kim, J. (2024). Evaluating YOLO Models for Efficient Crack Detection in Concrete Structures Using Transfer Learning. Buildings, 14(12), 3928. https://doi.org/10.3390/buildings14123928

Vancouver

Sohaib M, Arif M, Kim J. Evaluating YOLO Models for Efficient Crack Detection in Concrete Structures Using Transfer Learning. Buildings. 2024;14(12):3928. doi:10.3390/buildings14123928

BibTeX
@article{Sohaib2024,
  author = {Sohaib, Muhammad and Arif, Muzamal and Kim, Jong-Myon},
  title = {Evaluating YOLO Models for Efficient Crack Detection in Concrete Structures Using Transfer Learning},
  journal = {Buildings},
  year = {2024},
  volume = {14},
  number = {12},
  pages = {3928},
  doi = {10.3390/buildings14123928}
}
Kostenlos starten →