付龙,蔺小虎,娄宁,王玉杰,刘权

• 1:西安科技大学测绘学院
• 2:北京洛斯达科技发展有限公司

摘要(Abstract)：

针对无人机视角下道路缺陷检测中小尺度裂缝识别能力不足、下采样易丢失细节以及轻量化模型感受野有限、上下文建模能力弱等问题，提出一种改进的轻量化检测算法EWS-YOLOv8,以YOLOv8n为基础，在骨干和颈部网络中引入空间深度卷积，以增强低分辨率条件下对小目标的特征提取能力；设计C2f＿WTConv模块，通过小波变换扩展感受野并降低参数量与计算量；在骨干网络中嵌入高效多尺度注意力机制，利用跨空间学习捕捉多尺度信息以优化性能实验表明，相较于原始YOLOv8n模型，EWS-YOLOv8算法在检测精度与平均精度均值方面分别提升11%与3.4%,模型参数量降低24.3%,浮点运算数减少21%。研究旨在通过协同优化，在显著降低模型复杂度的同时，大幅提升道路微小缺陷的检测精度，满足边缘设备实时部署需求，适用于道路巡检系统中高精度与高效率并重的自动化检测任务。

关键词(KeyWords)： 道路缺陷检测;YOLOv8;空间深度卷积;C2f＿WTConv;高效多尺度注意力;小目标检测

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金青年项目(42201484);; 陕西省重点研发计划项目(2024GX-YBXM-294)

作者(Author): 付龙,蔺小虎,娄宁,王玉杰,刘权

DOI: 10.16251/j.cnki.1009-2307.2026.11.009

参考文献(References)：

• [1] SEKAR A,PERUMAL V.CFC-GAN:Forecasting road surface crack using forecasted crack generative adversarial network[J].IEEE Transactions on Intelligent Transportation Systems,2022,23(11):21378-21391.
• [2] 耿焕同，刘振宇，蒋骏，等.基于改进YOLOv8的嵌入式道路裂缝检测算法[J].计算机应用，2024,44(5):1613-1618.(GENG Huantong,LIU Zhenyu,JIANG Jun,et al.Embedded road crack detection algorithm based on improved YOLOv8[J].Journal of Computer Applications,2024,44(5):1613-1618.)
• [3] 王雪秋，高焕兵，郏泽萌.改进YOLOv8的道路缺陷检测算法[J].计算机工程与应用，2024,60(17):179-190.(WANG Xueqiu,GAO Huanbing,JIA Zemeng.Improved Road defect detection algorithm based on YOLOv8[J].Computer Engineering and Applications,2024,60(17):179-190.)
• [4] 戴激光，王梓晔，吴玉洁，等.面向农村违章建筑前置预警检测的IB-YOLO算法[J].测绘科学，2025,50(12):119-128.(DAI Jiguang,WANG Ziye,WU Yujie,et al.The IB-YOLO algorithm for pre-warning detection of rural illegal buildings[J].Science of Surveying and Mapping,2025,50(12):119-128.)
• [5] 徐光宪，唐桂芳，马飞.改进YOLOv8的无人机航拍道路损伤检测方法[J].测绘科学，2024,49(9):104-114.(XU Guangxian,TANG Guifang,MA Fei.Road damage detection method based on improved YOLOv8 in UAV aerial images[J].Science of Surveying and Mapping,2024,49(9):104-114.)
• [6] LUO Shipeng,ZHANG Yuxin,ZHANG Zeyu,et al.EPDD-YOLO:an efficient benchmark for pavement damage detection based on Mamba-YOLO[J].Measurement,2025,253:117638.
• [7] KHAN M W,OBAIDAT M S,MAHMOOD K,et al.Real-time road damage detection using an optimized YOLOv9s-fusion in IoT infrastructure[J].IEEE Internet of Things Journal,2025,12(11):17649-17660.
• [8] DONG Shuai,WANG Yunlong,CAO Jin,et al.Advanced lightweight deep learning vision framework for efficient pavement damage identification[J].Scientific Reports,2025,15:12966.
• [9] LI Rifa,GU Minming.Road crack detection method based on lightweight YOLOv8-LUAPD and UAV images[J].Measurement Science and Technology,2025,36(4):045411.
• [10] DU Yingjie,CHENG Qimin,LIU Xiaofeng,et al.Enhancing road maintenance through cyber-physical integration:The LEE-YOLO model for drone-assisted pavement crack detection[J].IEEE Transactions on Intelligent Transportation Systems,2025,26(9):14169-14178.
• [11] ARYA D,MAEDA H,GHOSH S K,et al.RDD2022:a multi-national image dataset for automatic road damage detection[J].Geoscience Data Journal,2024,11(4):846-862.
• [12] XIAO Xi,ZHANG Dianyan,HU Guangwu,et al.CNN-MHSA:a convolutional neural network and multi-head self-attention combined approach for detecting phishing websites[J].Neural Networks,2020,125:303-312.
• [13] XIE Xudong,XU Bing,CHEN Zhifei.Real-time fall attitude detection algorithm based on iRMB[J].Signal,Image and Video Processing,2024,19(2):156.
• [14] CAO Junxu,CHEN Qi,GUO Jun,et al.Attention-guided context feature pyramid network for object detection[EB/OL].[2026-03-05].https://arxiv.org/abs/2005.11475.
• [15] LI Chuyi,LI Lulu,JIANG Hongliang,et al.YOLOv6:a single-stage object detection framework for industrial applications[EB/OL].[2026-03-05].https://arxiv.org/abs/2209.02976.
• [16] VARGHESE R,M S.YOLOv8:a novel object detection algorithm with enhanced performance and robustness[C]//2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS).Chennai,India:IEEE,2024:1-6.
• [17] WANG C Y,YEH I H,MARK LIAO H Y.YOLOv9:Learning what you want toLearn using programmable gradient information[C]//Computer Vision - ECCV 2024.Cham:Springer,2025:1-21.
• [18] CHEN Hui,CHEN Kai,DING Guiguang,et al.YOLOv10:Real-time end-to-end object detection[C]//Advances in Neural Information Processing Systems 37.Vancouver,BC,Canada.Neural Information Processing Systems Foundation,Inc.(NeurIPS),2024:107984-108011.
• [19] KHANAM R,HUSSAIN M.YOLOv11:an overview of the key architectural enhancements[EB/OL].2024:arXiv:2410.17725.https://arxiv.org/abs/2410.17725.
• [20] TIAN Yunjie,YE Qixiang,DOERMANN D.YOLOv12:Attention-centric real-time object detectors[EB/OL].[2026-03-05].https://arxiv.org/abs/2502.12524.
• [21] LEI Mengqi,LI Siqi,WU Yihong,et al.YOLOv13:Real-time object detection with hypergraph-enhanced adaptive visual perception[EB/OL].[2026-03-05]https://arxiv.org/abs/2506.17733.
• [22] ZHU Xingkui,LYU Shuchang,WANG Xu,et al.TPH-YOLOv5:Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios[C]//2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW).Montreal,BC,Canada:IEEE,2021:2778-2788.
• [23] ZHAO Yian,LV Wenyu,XU Shangliang,et al.DETRs beat YOLOs on real-time object detection[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Seattle,WA,USA:IEEE,2024:16965-16974.
• [24] THIN JUN E L,THAM M L,KWAN B H.A comparative analysis of RT-DETR and YOLOv8 for urban zone aerial object detection[C]//2024 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS).Shah Alam,Malaysia:IEEE,2024:340-345.
• [25] DU D,ZHU P,WEN L,et al.VisDrone-DET2019:The vision meets drone object detection in image challenge results[C]//Proceedings of the IEEE/CVF international conference on computer vision workshops.2019:213-226.