Crack-MsCGA: A Deep Learning Network with Multi-Scale Attention for Pavement Crack Detection
Sensors · 12.04.2025 · 25 (8) · 2446
DOI: 10.3390/s25082446 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 12.04.2025. Verfasst von Guoxi Liu und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650224, China. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Research and Development Program of Yunnan Province, National Natural Science Foundation, Yunnan Provincial Revitalization Talents Support Plan.
Abstract (Originalsprache)
Pavement crack detection is crucial for ensuring road safety and reducing maintenance costs. Existing methods typically use convolutional neural networks (CNNs) to extract multi-level features from pavement images and employ attention mechanisms to enhance global features. However, the fusion of low-level features introduces substantial interference, leading to low detection accuracy for small-scale cracks with subtle local structures and varying global morphologies. In this paper, we propose a computationally efficient deep learning network with CNNs and multi-scale attention for multi-scale crack detection, named Crack-MsCGA. In this network, we avoid fusing low-level features to reduce noise interference. Then, we propose a multi-scale attention mechanism (MsCGA) to learn local detail features and global features from high-level features, compensating for the reduced detailed information. Specifically, first, MsCGA employs local window attention to learn short-range dependencies, aggregating local features within each window. Second, it applies a cascaded group attention mechanism to learn long-range dependencies, extracting global features across the entire image. Finally, it uses a multi-scale attention fusion strategy based on Mixed Local Channel Attention (MLCA) selectively to fuse local features and global features of pavement cracks. Compared with five existing methods, it improves the AP@50 by 11.3% for small-scale, 8.1% for medium-scale, and 5.9% for large-scale detection over the state-of-the-art methods in the DH807 dataset.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650224, China.
- School of Computer Science and Engineering, South China University of Technology, Guangzhou 510641, China.
ORCID-Profile: Guoxi Liu, Fei Dai, Guozhi Liu, Lecheng Li
Förderung
- Research and Development Program of Yunnan Province
- National Natural Science Foundation
- Yunnan Provincial Revitalization Talents Support Plan
- Science and Technology Youth Lift Talents of Yunnan Province, China, Key Industry Science and Technology Projects for University Services in Yunnan Province
- Yunnan Provincial Department of Education Science Research Fund
- Yunnan Fundamental Research Projects
- Open Fund Project of Key Laboratory of the Ministry of Education
- Research and Development Program of Yunnan Province, China
Zitieren
APA
Liu, G., Wu, X., Dai, F., Liu, G., Li, L., & Huang, B. (2025). Crack-MsCGA: A Deep Learning Network with Multi-Scale Attention for Pavement Crack Detection. Sensors, 25(8), 2446. https://doi.org/10.3390/s25082446
Vancouver
Liu G, Wu X, Dai F, Liu G, Li L, Huang B. Crack-MsCGA: A Deep Learning Network with Multi-Scale Attention for Pavement Crack Detection. Sensors. 2025;25(8):2446. doi:10.3390/s25082446
BibTeX
@article{Liu2025,
author = {Liu, Guoxi and Wu, Xiaojing and Dai, Fei and Liu, Guozhi and Li, Lecheng and Huang, Bi},
title = {Crack-MsCGA: A Deep Learning Network with Multi-Scale Attention for Pavement Crack Detection},
journal = {Sensors},
year = {2025},
volume = {25},
number = {8},
pages = {2446},
doi = {10.3390/s25082446}
}
EUVIMED