.png){kind=icon}
.png){kind=icon}
Return
An enhanced YOLO-based deep learning framework for automated structural defect inspection in complex backgrounds
.png){kind=icon}
PRE
AI
Z
D
Y
D
Z
DOI：10.1016/j.engstruct.2026.123483.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• An integrated automated framework enables bridge defect assessment, linking diagnostics with on-site inspection. • A lightweight YOLO-based detection model developed to enhance feature extraction and domain adaptation. • A dual-branch attention module with large-kernel separable convolutions captures both cracks and spalling effectively. • The framework and PySide6-based interface enable accurate defect localization and bridge maintenance.
Journal
IF:
6.4
Papers:
2.1W
Citations:
8.7W
Organization
Cited Papers