Please wait a minute...
 
隧道与地下工程灾害防治
  研究论文 本期目录 | 过刊浏览 | 高级检索 |
隧道施工环境中障碍物的轻量化目标检测算法
吴江涛1, 李英杰2
(1.中国水利水电第四工程局有限公司,青海 西宁,810007;2.山东大学齐鲁交通学院,山东 济南,250002)
The lightweight object detection algorithm for obstacles in tunnel construction environments#br#
WU Jiangtao1, LI Yingjie2
(1.Sinohydro Enginnering Bureau 4 Co., Ltd., Xining 810007, Qinghai, China;2.School of Qilu Transportation, Shandong University, Jinan 250002, Shandong,  China)
下载:  PDF (1925KB) 
输出:  BibTeX | EndNote (RIS)      
摘要 由于隧道施工现场事故频发,通过机器人代替人工进行安全巡检可有效保障工作人员的生命安全。然而,目前该领域的机器人大多需要人工操控,缺乏自主避障能力。针对隧道施工环境下的障碍物检测方法,提出利用改进YOLOv5进行障碍物检测的轻量化模型。首先构建隧道场景下的障碍物数据集;其次修改骨干网络为轻量级Shufflenet v2网络,并将其中的激活函数修改为SiLU函数,以提高检测速度降低计算量;然后引入坐标注意力机制,增强网络学习特征的表达能力;最后将颈部卷积块修改为GSConv,在减小模型计算量的同时提升算法的检测精度。研究结果表明:基于本研究所构建的数据集上进行对比试验,该方法的检测速度较YOLOv5-n原始算法提升了57%,减少了模型对硬件的需求。
服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
吴江涛
李英杰
关键词:  深度学习  目标检测  改进YOLOv5  障碍物检测    
Abstract: Due to the frequent accidents at the tunnel construction site, the use of robots instead of manual safety inspections can effectively protect the life safety of staff. However, most of the robots in this field currently require manual control and lack autonomous obstacle avoidance capabilities. Aiming at the obstacle detection method in tunnel construction environment, a lightweight model using improved YOLOv5 for obstacle detection was proposed. Firstly, the obstacle dataset in the tunnel scenario was constructed. Secondly, the backbone network was modified to a lightweight Shufflenet v2 network, and the activation function is modified to the SiLU function to improve the detection speed and reduce the amount of computation. Next, the coordinate attention mechanism was incorporated to improve the network's ability to learn and represent features. Finally, the neck convolution block was modified to GSConv, which reduced the calculation amount of the model and improved the detection accuracy of the algorithm. Comparative experiments on the dataset constructed in this research showed that the detection speed of the proposed method was increased by about 57% compared with the original YOLOv5-n algorithm, which greatly reduces the hardware requirements of the model.
Key words:  deep learning    object detection    improvement of YOLOv5    obstacle detection
收稿日期:  2024-12-19      修回日期:  2025-02-26      发布日期:  2025-02-27     
中图分类号:  TU443  
基金资助: 山东铁投集团科技计划资助项目(TTKJ2021-06)
作者简介:  吴江涛(1982—)男 ,陕西西安人,高级工程师,主要研究方向为高速铁路施工。E-mail:505962421@qq.com
引用本文:    
吴江涛, 李英杰. 隧道施工环境中障碍物的轻量化目标检测算法[J]. 隧道与地下工程灾害防治, .
WU Jiangtao, LI Yingjie. The lightweight object detection algorithm for obstacles in tunnel construction environments#br#. Hazard Control in Tunnelling and Underground Engineering, 0, (): 1-.
链接本文:  
[1] 高贤成. 基于ConDenseNet架构煤岩破坏识别模型及其优化研究[J]. 隧道与地下工程灾害防治, 2024, 6(4): 90-98.
[2] 方恩权, 王耀东, 李星言, 马农杰. 地铁盾构隧道渗漏水病害图像识别算法[J]. 隧道与地下工程灾害防治, 2022, 4(4): 28-33.
No Suggested Reading articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
网站版权 © 《隧道与地下工程灾害防治》编辑部
地址:山东省济南市山大南路27号山东大学中心校区明德楼B733《隧道与地下工程灾害防治》编辑部, 邮编:250100, 电话:0531-88366735, E-mail:tunnel@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn