A Transformer-Based Pavement Crack Segmentation Model with Local Perception and Auxiliary Convolution Layers
Electronics · 15.07.2025 · 14 (14) · 2834
DOI: 10.3390/electronics14142834 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 15.07.2025. Verfasst von Yi Zhu und 2 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Industrial Technology Development Project, Special Project of Technological Innovation and Guidance in Shaanxi Province, Natural Science Foundation in Shaanxi Province.
Abstract (Originalsprache)
Crack detection in complex pavement scenarios remains challenging due to the sparse small-target features and computational inefficiency of existing methods. To address these limitations, this study proposes an enhanced architecture based on Mask2Former. The framework integrates two key innovations. A Local Perception Module (LPM) reconstructs geometric topological relationships through a Sequence-Space Dynamic Transformation Mechanism (DS2M), enhancing neighborhood feature extraction via depthwise separable convolutions. Simultaneously, an Auxiliary Convolutional Layer (ACL) combines lightweight residual convolutions with shallow high-resolution features, preserving critical edge details through channel attention weighting. Experimental evaluations demonstrate the model’s superior performance, achieving improvements of 3.2% in mIoU and 2.7% in mAcc compared to baseline methods, while maintaining computational efficiency with only 12.8 GFLOPs. These results validate the effectiveness of geometric relationship modeling and hierarchical feature fusion for pavement crack detection, suggesting practical potential for infrastructure maintenance systems. The proposed approach balances precision and efficiency, offering a viable solution for real-world applications with complex crack patterns and hardware constraints.
Förderung
- Industrial Technology Development Project
- Special Project of Technological Innovation and Guidance in Shaanxi Province
- Natural Science Foundation in Shaanxi Province
- Science and Technology Program in Xi’an city
- Key research and development plan of Shaanxi Province
Zitieren
APA
Zhu, Y., Cao, T., & Yang, Y. (2025). A Transformer-Based Pavement Crack Segmentation Model with Local Perception and Auxiliary Convolution Layers. Electronics, 14(14), 2834. https://doi.org/10.3390/electronics14142834
Vancouver
Zhu Y, Cao T, Yang Y. A Transformer-Based Pavement Crack Segmentation Model with Local Perception and Auxiliary Convolution Layers. Electronics. 2025;14(14):2834. doi:10.3390/electronics14142834
BibTeX
@article{Zhu2025,
author = {Zhu, Yi and Cao, Ting and Yang, Yiqing},
title = {A Transformer-Based Pavement Crack Segmentation Model with Local Perception and Auxiliary Convolution Layers},
journal = {Electronics},
year = {2025},
volume = {14},
number = {14},
pages = {2834},
doi = {10.3390/electronics14142834}
}
EUVIMED