Urban road surface crack detection based on U-net and ResNeXt network

Jun Qiao, Huabing Wang, Zidong Zhou, Yunwei Meng and Minghui Gong

PLOS ONE, 2026, vol. 21, issue 4, 1-24

Abstract: With the continuous increase in urban road usage, various cracks often appear on the road surface, which may pose a threat to traffic safety. Presently, road inspection is still primarily limited to manual methods, which suffer from low efficiency, limited accuracy, and subjective judgment. To enhance the efficiency of road crack detection, the paper designs an innovative detection technology that fuses U-net and ResNeXt networks. The results showed that the proposed method achieved superior detection performance on horizontal and vertical cracks. While its recognition and classification capabilities for other types of cracks and block cracks need improvement, it still demonstrated significant overall classification performance. Compared with numerous detection methods, the performance of the proposed method was notably superior. The peak memory efficiency of the video memory of this method is controlled within 2.1GB. This indicates that in practical applications, the proposed method can provide accurate information on road surface cracks, making it easier for workers to take corresponding remedial measures. In summary, the proposed urban road surface crack detection method can be integrated into intelligent transportation systems, providing technical support for real-time monitoring and predictive maintenance of road conditions.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0347145 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 47145&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0347145

DOI: 10.1371/journal.pone.0347145

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().