自适应特征的轻量化路面裂缝检测方法

打开文本图片集

关键词：路面裂缝检测；注意力机制；轻量化；动态采样金字塔；YOLOv8中图分类号：TP394.1 文献标识码：Adoi：10.37188/0PE.20263402.0336 CSTR：32169.14.OPE.20263402.0336

Lightweight road crack detection method with adaptive features

LIUYuanyuan，ZHUKai，GU Zhihui，YUE Meng，WANG Jingzhi，ZHULu （School ofInformation andSoftware Engineering，East China Jiaotong Unirversity，Nanchang 33OOl3，China） *Correspondingauthor， E -mail： luyuanwanwan@l63. com

Abstract：To address the issues of complex crack morphologies，environmental interference，and the imbalance between detection acuracy and model lightweight requirements in road surface inspection，this paper proposed a lightweight road crack detection method with adaptive feature extraction. First，a Crack Efficient Attention （CEA）module was designed based on the slender shape and large span characteristics of cracks，compressing feature dimensions to capture long-distance spatial dependencies. Second，a Dynamic Sampling Feature Pyramid Network （DSFPN） was constructed for adaptive sampling and target feature extraction，enhancing representation capability for heterogeneous crack features. Third，the HGNet_GS lightweight backbone network was improved，and a CEA Group Head （CGHead）was proposed，significantly reducing computational redundancy； the PIoU （Powerful IoU） loss function was adopted to solve anchor box expansion problems and improve convergence speed for smallmodels. Additionally，a civilian road defect dataset containing 2 985 images under various lighting conditions Was established to validate model generalization. Experimental results show that compared with the baseline YOLOv8n model，the proposed method reduces parameters and computational cost by 50% and 52% ，respectively；on the selfbuiltdataset，mAP5O and mAP95 increaseby 5.4% and 4.1% ；on the public RDD2O22dataset，these metrics improve by 2.1% and 1.5% . The model has been deployed on edge devices and verified through engineering tests，demonstrating its capability to meet practical requirements for lightweight road crack detection and providing a technical solution for automated road maintenance systems.

Key words： pavement crack detection； attention mechanism； lightweight; dynamic pyramid; YOLOv8

1引言

道路作为交通系统的基石，是推动经济发展与城市化进程不可或缺的关键要素。（剩余20762字）