基于改进YOLOv5的路面病害检测方法  

Pavement Disease Detection Method Based on Improved YOLOv5

作　　者：刘鹏宇 袁静[1,2,3] 高倩 陈善继[4] LIU Pengyu;YUAN Jing;GAO Qian;CHEN Shanji(Faculty of Information Technology,Beijing University of Technology,Beijing 100124,China;Beijing Laboratory of Advanced Information Networks,Beijing 100124,China;Beijing Key Laboratory of Computational Intelligence and Intelligent System,Beijing University of Technology,Beijing 100124,China;School of Physics and Electronic Information Engineering,Qinghai Minzu University,Xining 810007,China)

机构地区：[1]北京工业大学信息学部,北京100124 [2]先进信息网络北京实验室,北京100124 [3]北京工业大学计算智能与智能系统北京市重点实验室,北京100124 [4]青海民族大学物理与电子信息工程学院,西宁81000

出　　处：《北京工业大学学报》2025年第5期552-559,共8页Journal of Beijing University of Technology

基　　金：青海省应用基础研究计划资助项目(2021-ZJ-704)。

摘　　要：针对目前道路病害检测数据集种类较少、检测场景单一,以及现有基于深度学习的路面病害检测方法难以应对复杂环境干扰、模型由于体积较大难以部署等问题,建立一个多种类、面向多种场景类型的路面病害检测数据集,以弥补现有数据集的不足,并且提出基于改进YOLOv5的路面病害检测方法。该方法通过融合注意力机制和轻量化结构组件在提升模型检测精度的同时降低参数量,实现了在多种干扰背景下对裂缝和坑槽路面损坏的检测和准确识别,有效改善了上述不足。实验结果表明,提出的方法在构建的路面病害数据集上检测平均精度均值达到93.3%,具有较高的检测精度,模型参数量仅为6.7×10^(6)左右,大大降低了部署成本。Currently,there is a scarcity of road disease detection data sets,single detection scenarios,the existing road disease detection methods based on deep learning are difficult to deal with complex environmental interference,and the model size is too large to deploy.A multi-type and scenario oriented pavement disease detection data set was established to make up for the shortcomings of existing data sets.Furthermore,a pavement disease detection method based on improved YOLOv5 was proposed.This method integrated an attention mechanism and lightweight structural components to improve the model detection accuracy while reducing the number of parameters,achieving the detection and accurate identification of cracks and potholes pavement damage under various interference backgrounds,and effectively improving the aforementioned deficiencies.Results show that the proposed method has a high mean average precison of 93.3%on the constructed pavement disease data set,and the number of model parameters is only about 6.7×10^(6),which significantly reduces the deployment cost.

关 键 词：公路养护 路面病害 深度学习 YOLOv5 注意力机制 轻量化 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]