1 College of Geodesy and Feomatics, Shandong University of Science and Technology, Qingdao 266590, China
2 Research Institute of Highway Ministry of Transport, Beijing 100088, China
3 Innovation Research Institute of Shandong Expressway Group Co., Jinan 250102, China
Show less
文章历史+
Received
Published
Revised
2025-03-13
2026-02-28
2025-12-01
Online Date
2026-05-30
PDF (18856K)
摘要
为解决现有道路裂缝检测模型精度低、易漏检误检等问题,提出了一种基于混合局部通道注意力机制的YOLOv10n道路裂缝检测模型。针对道路裂缝细长且不规则的形状特点,首先在模型特征提取网络引入混合局部通道注意力机制(mixed local channel attention, MLCA),增强模型对道路裂缝特征的提取能力;其次,采用重参数化泛化特征金字塔网络(reparameterized generalized feature pyramid network,RepGFPN)优化原始颈部网络,充分融合多尺度下的裂缝特征信息;最后使用Focaler-IoU替换CIoU损失函数,调整模型训练不同裂缝样本的权重,加快收敛速度。在RDD2022_China数据集上的实验结果表明,改进后的模型相较于原始YOLOv10n模型检测准确率提升4.4%,平均精度均值(mean average precision, mAP)提高2.9%。与其他主流目标检测模型相比,改进后的模型在准确率、召回率和计算成本等方面均展现出最佳性能,验证了本文方法在道路裂缝检测任务中的有效性和优越性。
Abstract
In order to solve the problems of low accuracy of the existing road crack detection model and easy to miss and misdetect, a road crack detection model based on MLCA yolov10n was proposed. In view of the slender and irregular shape of road cracks, MLCA (mixed local channel attention) was introduced into the model feature extraction network to enhance the extraction capability of the model for road crack features. Secondly, RepGFPN (reparameterized generalized feature pyramid network) was used to optimise the original neck network to fully integrate the crack feature information at multiple scales. Finally, the Focaler-IoU was used to replace the CIoU loss function and adjust the weights of different crack samples for model training to speed up the convergence. Experimental results on the RDD2022_China dataset show that the improved model improves detection accuracy by 4.4% and mAP(mean average precision) by 2.9% compared to the original YOLOv10n model. Compared with other mainstream target detection models, the improved model demonstrates the best performance in terms of accuracy, recall and computational cost, which validates the effectiveness and superiority of this paper’s method in road crack detection tasks.
LIU Ru-fei, LIU Zhao-xu, REN Hong-wei, SU Zhan-wen, CONG Bo-ri.
Road Crack Detection Model Based on MLCA[J].
Science Technology and Engineering, 2026, 26(6): 2484-2492 DOI:10.12404/j.issn.1671-1815.2501712
ZhouY, GuoX M, HouF J, et al. Review of Intelligent Road Defects Detection Technology[J]. Sustainability, 2022, 14(10). DOI:10.3390/su14106306.
[2]
QuanY W, SunJ, ZhangY, et al. The method of the road surface crack detection by the improved OTSU threshold[C]// International Conference on Mechatronics and Automation. New York: IEEE,2019: 1615-1620.
[3]
YaacoubC, DaouR A Z. Fractional order sobel edge detector[C]//Ninth International Conference on Image Processing Theory, Tools and Applications. New York: IEEE, 2019: 1-5.
[4]
ChenF C, JahanshahiM R, WuR T, et al. A texture-based video processing methodology using Bayesian data fusion for autonomous crack detection on metallic surfaces[J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(4): 271-287.
[5]
ShenG. Road crack detection based on video image processing[C]//3rd International Conference on Systems and Informatics. New York: IEEE, 2016: 912-917.
ZhangXiaohua, LiXiaolong, AiJinquan, et al. Road crack detection algorithm based on improved Mask R-CNN in complex backgrounds[J]. Beijing Surveying and Mapping, 2024, 38(3): 431-436.
[10]
SuH, KamandaD B, GuoC, et al. Enhanced YOLOv3 for precise detection of apparent damage on bridges amidst complex backgrounds[J]. Scientific Reports, 2024, 14(1): 8627.
[11]
HanZ Q, JunGand YuQ. RDDS: road disease detection model based on YOLOv5[C]// IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems. New York: IEEE, 2024: 511-516.
LiuC, YaoY S, LiJ, et al. Research on lightweight GPR road surface disease image recognition and data expansion algorithm based on YOLO and GAN[J]. Case Studies in Construction Materials, 2024, 20.DOI: 10.1016/j.cscm.2023.e02779.
[14]
YangX L, ShengH, GengN, et al. Research on road disease detection algorithm based on improved YOLOv5s[C]// 6th International Conference on Electronics and Communication, Network and Computer Technology. New York: IEEE, 2024: 305-309.
[15]
MaM Y, SongS C, LiH X, et al. Research on road crack detection in complex background based on MFF-YOLO[C]// 4th International Conference on Neural Networks, Information and Communication Engineering. New York: IEEE, 2024: 919-924.
ZhuWeigang, WangLun, ChenTian, et al. UAV image road crack detection method based on improved YOLOv5[J]. Bulletin of Surveying and Mapping, 2024(3): 173-178.