MSP U-Net: Crack Segmentation for Low-Resolution Images Based on Multi-Scale Parallel Attention U-Net
Applied Sciences · 11.12.2024 · 14 (24) · 11541
DOI: 10.3390/app142411541 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 11.12.2024. Verfasst von Joon-Hyeok Kim und 3 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Education.
Abstract (Originalsprache)
As the expected lifespans of structures and road approaches, as well as the importance of road maintenance, increase globally, safety inspections have emerged as a crucial task. Nonetheless, the existing crack detection models focus on multi-scale feature loss and performance degradation in learning various types of cracks. We propose the Multi-Scale Parallel Attention U-Net (MSP U-Net) as a network designed for low-resolution images that considers the irregular characteristics of cracks. MSP U-Net applies a large receptive field flock to an attention U-Net, minimizing feature loss across multiple scales. Using the Crack500 dataset, our network achieved a mean intersection of union (mIoU) of 0.7752, outperforming the existing methods on low-resolution images.
Förderung
- Ministry of Education
Zitieren
APA
Kim, J., Noh, J., Jang, J., & Yang, H. (2024). MSP U-Net: Crack Segmentation for Low-Resolution Images Based on Multi-Scale Parallel Attention U-Net. Applied Sciences, 14(24), 11541. https://doi.org/10.3390/app142411541
Vancouver
Kim J, Noh J, Jang J, Yang H. MSP U-Net: Crack Segmentation for Low-Resolution Images Based on Multi-Scale Parallel Attention U-Net. Applied Sciences. 2024;14(24):11541. doi:10.3390/app142411541
BibTeX
@article{Kim2024,
author = {Kim, Joon-Hyeok and Noh, Ju-Hyeon and Jang, Jun-Young and Yang, Hee-Deok},
title = {MSP U-Net: Crack Segmentation for Low-Resolution Images Based on Multi-Scale Parallel Attention U-Net},
journal = {Applied Sciences},
year = {2024},
volume = {14},
number = {24},
pages = {11541},
doi = {10.3390/app142411541}
}
EUVIMED