Self-supervised multi-scale transformer with Attention-Guided Fusion for efficient crack detection
Automation in Construction · 2026-01 · 181 · 106591
DOI: 10.1016/j.autcon.2025.106591 ↗Kurz gefasst
Fachartikel in der Zeitschrift Automation in Construction, 2026-01. Verfasst von Blessing Agyei Kyem und 4 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 67 Literaturangaben. Gefördert von: North Dakota State University, University Transportation Centers.
Abstract (Originalsprache)
Pavement crack detection has long depended on costly and time-intensive pixel-level annotations, which limit its scalability for large-scale infrastructure monitoring. To overcome this barrier, this paper examines the feasibility of achieving effective pixel-level crack segmentation entirely without manual annotations. Building on this objective, a fully self-supervised framework, Crack-Segmenter, is developed, integrating three complementary modules: the Scale-Adaptive Embedder (SAE) for robust multi-scale feature extraction, the Directional Attention Transformer (DAT) for maintaining linear crack continuity, and the Attention-Guided Fusion (AGF) module for adaptive feature integration. Through evaluations on ten public datasets, Crack-Segmenter consistently outperforms 13 state-of-the-art supervised methods across all major metrics, including mean Intersection over Union (mIoU), Dice score, XOR, and Hausdorff Distance (HD). These findings demonstrate that annotation-free crack detection is not only feasible but also superior, enabling transportation agencies and infrastructure managers to conduct scalable and cost-effective monitoring. This work advances self-supervised learning and motivates pavement cracks detection research.
Förderung
- North Dakota State University
- University Transportation Centers
Zitieren
APA
Kyem, B. A., Asamoah, J. K., Denteh, E., Danyo, A., & Aboah, A. (2026). Self-supervised multi-scale transformer with Attention-Guided Fusion for efficient crack detection. Automation in Construction, 181, 106591. https://doi.org/10.1016/j.autcon.2025.106591
Vancouver
Kyem BA, Asamoah JK, Denteh E, Danyo A, Aboah A. Self-supervised multi-scale transformer with Attention-Guided Fusion for efficient crack detection. Automation in Construction. 2026;181:106591. doi:10.1016/j.autcon.2025.106591
BibTeX
@article{Kyem2026,
author = {Kyem, Blessing Agyei and Asamoah, Joshua Kofi and Denteh, Eugene and Danyo, Andrews and Aboah, Armstrong},
title = {Self-supervised multi-scale transformer with Attention-Guided Fusion for efficient crack detection},
journal = {Automation in Construction},
year = {2026},
volume = {181},
pages = {106591},
doi = {10.1016/j.autcon.2025.106591}
}
EUVIMED