一种基于YOLO11的路面裂缝分割算法
By using a YOLO11-based pavement crack segmentation algorithm and leveraging multi-scale feature enhancement and polarization self-attention mechanisms, the accuracy and robustness issues of crack detection in complex environments are addressed, achieving efficient and accurate crack segmentation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIAN UNIV OF TECH
- Filing Date
- 2025-08-19
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for detecting road surface cracks struggle to accurately distinguish between cracks and background in complex environments, leading to incomplete segmentation or missed detections. Furthermore, they perform poorly in detecting minute cracks, failing to meet the demands of modern road maintenance for high precision and efficiency.
A road crack segmentation algorithm based on YOLO11 is adopted. Through multi-scale feature enhancement and polarization self-attention mechanism, polarization feature weights of channel and spatial dimensions are constructed to enhance the multi-dimensional focusing ability of crack detection and improve detection accuracy and robustness.
It improves the accuracy and robustness of crack detection in complex environments, better identifies multi-scale cracks, reduces background interference, and achieves efficient and accurate crack segmentation.
Smart Images
Figure CN121236375B_ABST