Open access
Aug 2026
Semantic segmentation and quantitative analysis of tunnel cracks and water leakage using a TransUNet framework
This study proposes an end-to-end semantic segmentation framework based on TransUNet that significantly enhances the characterization of multi-scale defects and boundary features and meets the requirements of offline inspection and near-real-time applications.
Xinjian Li, Qiaofeng Liu, Gang Yan et al.
· PLoS ONE · 0 citations