摘要
随着我国桥梁建设规模的不断扩大和桥梁设施的日益老化,部分桥梁可能在未来20年内面临结构劣化问题,因此检测成为保障桥梁安全运行的重要环节。传统的桥梁检测方法受人工操作和设备限制,存在效率低、成本高、精度不足、无法考证等问题。基于无人机的桥梁检测可部分替代传统的人工桥梁检测作业,有利于桥梁检测智能化、自动化。为实现无人机桥梁检测的自动化,提出基于深度学习的桥梁图像自动分类方法,并通过模型训练验证该方法的有效性。
With the expanding scale of bridge construction and the increasing aging of bridge facilities in China,some bridges may face structural deterioration in the next 20 years,so inspection has become an important link to ensure the safe operation of bridges.Traditional bridge inspection methods are limited by manual operation and equipment,and have problems such as low efficiency,high cost,lack of accuracy and unverifiable.Based on unmanned aerial vehicle(UAV)bridge inspection can partially replace the traditional manual bridge inspection operation,which is conducive to intelligent and automated bridge inspection.In order to realize the automation of UAV bridge inspection,an automatic bridge image classification method based on deep learning is proposed,and the effectiveness of the method is verified by model training.
作者
赵荣欣
余威镭
叶从周
王枫
吴华勇
周子杰
ZHAO Rongxin;YU Weilei;YE Congzhou;WANG Feng;WU Huayong;ZHOU Zijie(Shanghai Key Laboratory of Engineering Structure Safety,Shanghai Research Institute of Building Sciences Co.,Ltd.,Shanghai 200032,China)
出处
《施工技术（中英文）》
CAS
2023年第9期7-10,共4页
Construction Technology
基金
上海市科学技术委员会优秀技术带头人项目(20XD1432400)
上海建科集团科研项目(KY10000038.20210005)。
关键词
桥梁
深度学习
卷积神经网络
Resnet网络
无人机
图像分类
检测
巡检
bridges
deep learning
convolutional neural network(CNN)
Resnet network
unmanned aerial vehicle(UAV)
image classification
detection
inspection
