EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Real-Time Mobile Robotic System for Crack Detection in Construction Using Two-Stage Deep Learning

Emmanuella Ogun, Yong Ann Voeurn iD, Doyun Lee iD

Sensors · 13.01.2026 · 26 (2) · 530

DOI: 10.3390/s26020530 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 13.01.2026. Verfasst von Emmanuella Ogun und 2 weiteren Autor:innen. Beteiligte Einrichtung: Mechanical Engineering Department, Georgia Southern University, Statesboro, GA 30460, USA. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The deterioration of civil infrastructure poses a significant threat to public safety, yet conventional manual inspections remain subjective, labor-intensive, and constrained by accessibility. To address these challenges, this paper presents a real-time robotic inspection system that integrates deep learning perception and autonomous navigation. The proposed framework employs a two-stage neural network: a U-Net for initial segmentation followed by a Pix2Pix conditional generative adversarial network (GAN) that utilizes adversarial residual learning to refine boundary accuracy and suppress false positives. When deployed on an Unmanned Ground Vehicle (UGV) equipped with an RGB-D camera and LiDAR, this framework enables simultaneous automated crack detection and collision-free autonomous navigation. Evaluated on the CrackSeg9k dataset, the two-stage model achieved a mean Intersection over Union (mIoU) of 73.9 ± 0.6% and an F1-score of 76.4 ± 0.3%. Beyond benchmark testing, the robotic system was further validated through simulation, laboratory experiments, and real-world campus hallway tests, successfully detecting micro-cracks as narrow as 0.3 mm. Collectively, these results demonstrate the system’s potential for robust, autonomous, and field-deployable infrastructure inspection.

Schlagwörter

Stichwörter der Autor:innen

Pix2PixROS 2SLAMU-Netautonomous inspectioncrack detectiondeep learningmobile roboticssemantic segmentationstructural health monitoring

Autor:innen und Institutionen

  • Mechanical Engineering Department, Georgia Southern University, Statesboro, GA 30460, USA.
  • Civil Engineering and Construction Department, Georgia Southern University, Statesboro, GA 30460, USA.

ORCID-Profile: Yong Ann Voeurn, Doyun Lee

Zitieren

APA

Ogun, E., Voeurn, Y. A., & Lee, D. (2026). A Real-Time Mobile Robotic System for Crack Detection in Construction Using Two-Stage Deep Learning. Sensors, 26(2), 530. https://doi.org/10.3390/s26020530

Vancouver

Ogun E, Voeurn YA, Lee D. A Real-Time Mobile Robotic System for Crack Detection in Construction Using Two-Stage Deep Learning. Sensors. 2026;26(2):530. doi:10.3390/s26020530

BibTeX
@article{Ogun2026,
  author = {Ogun, Emmanuella and Voeurn, Yong Ann and Lee, Doyun},
  title = {A Real-Time Mobile Robotic System for Crack Detection in Construction Using Two-Stage Deep Learning},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {2},
  pages = {530},
  doi = {10.3390/s26020530}
}
Kostenlos starten →