[1]冀承泽,贾立新,李荆晖.基于改进 YOLOv5s 的两种输电杆塔缺陷检测研究[J].计算机技术与发展,2024,34(02):180-185.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 026]
　JI Cheng-ze,JIA Li-xin,LI Jing-hui.Research on Two Types of Defect Detection of Transmission Tower Based on Improved YOLOv5s[J].,2024,34(02):180-185.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 026]
点击复制

基于改进 YOLOv5s 的两种输电杆塔缺陷检测研究()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
34
期数:
2024年02期
页码:
180-185
栏目:
新型计算应用系统
出版日期:
2024-02-10

文章信息/Info

Title:
Research on Two Types of Defect Detection of Transmission Tower Based on Improved YOLOv5s
文章编号:
1673-629X(2024)02-0180-06
作者:
冀承泽; 贾立新; 李荆晖
西安交通大学 电气工程学院,陕西 西安 710049
Author(s):
JI Cheng-ze; JIA Li-xin; LI Jing-hui
School of Electrical Engineering,Xi’ an Jiaotong University,Xi’ an 710049,China
关键词:
YOLOv5s; 输电杆塔; 缺陷检测; 深度网络; 损失函数
Keywords:
YOLOv5s; transmission towers; defect detection; deep network; loss function
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2024. 02. 026
摘要:
国内的电力事业发展迅速,输电杆塔的缺陷检测与修复是保证电网安全运行的关键技术手段。 当前主要是人为识别输电杆塔的缺陷,工作负担巨大。 故以 YOLOv5s 网络为基础,提出一种改进 YOLOv5s 目标检测算法,提升检测效率。在基础模型上引入 Focal-EIoU 损失函数,提升模型收敛速度与精度;在卷积层引入 Hardswish 激活函数,提高模型的表达能力,查准率得到提升;上调算法推理的置信度阈值 conf-thres,减少模型推理的误检情况,提升模型正检率。 另外在研究中尝试融入注意力机制提升网络特征提取能力,但效果不好,故舍弃此改进策略。 实验结果表明,改进模型的各项指标均获得了提升,查准率由 92. 96% 提升至 95. 02% ,上涨了 2. 06 百分点;查全率由 87. 36% 提升到了 87. 38% ;mAP@ . 5 ：mAP@ . 5 ：. 95(0. 1 ：0. 9) 由 0. 644 3 提升至 0. 648 1,上涨了 0. 38 百分点;模型检测速度 FPS 提高了 4. 4。
Abstract:
With the rapid development of domestic electric power industry,the defect detection and repair of transmission towers are keytechnical means to ensure the safe operation of the power grid. At present, it is mainly to identify the defects of transmission towermanually,and the work burden is huge. Therefore,we propose an improved YOLOv5s object detection algorithm based on the YOLOv5snetwork to improve the efficiency of detection. Focal-EIoU loss function is introduced into the basic model to improve the convergencespeed and accuracy of the model. The Hardswish activation function is introduced into the convolution layer to improve the expressionability of the model and the precision. The confidence threshold conf-thres of the algorithm reasoning is increased to reduce the false detection of model reasoning and improve the positive detection rate of the model. In addition,in the research,we tried to integrate attentionmechanism to improve the ability of network feature extraction,but the effect was not ideal,so we abandoned this improvement strategy.The experimental results show that all indicators of the improved model have been improved,with a precision increase of 2. 06 percentfrom 92. 96% to 95. 02% ; the recall rate has increased from 87. 36% to 87. 38% ;?mAP@ . 5 ：mAP@ . 5 ：. 95(0. 1 ：0. 9) increasedfrom 0. 644 3 to 0. 648 1,an increase of 0. 38 percent;the model detection speed FPS has been improved by 4. 4.

相似文献/References:

[1]赵崇志,卢 俊,田 源,等.基于 Yolov5s 的作业现场阀门规范操作识别方法[J].计算机技术与发展,2022,32(05):216.[doi:10. 3969 / j. issn. 1673-629X. 2022. 05. 036]
　ZHAO Chong-zhi,LU Jun,TIAN Yuan,et al.Yolov5s-based Method for Identifying Valve Specification Operations at Job Sites[J].,2022,32(02):216.[doi:10. 3969 / j. issn. 1673-629X. 2022. 05. 036]
[2]刘鸿瑜,袁国武.基于改进 YOLOv5s 的烟支外观缺陷检测方法[J].计算机技术与发展,2022,32(08):161.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 026]
　LIU Hong-yu,YUAN Guo-wu.Detection of Cigarette Appearance Defects Based on Improved YOLOv5s[J].,2022,32(02):161.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 026]
[3]朱昭云,张 波,叶昭良,等.基于 YOLOv5s 的海上风电设施检测与预警评估[J].计算机技术与发展,2023,33(04):182.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 027]
　ZHU Zhao-yun,ZHANG Bo,YE Zhao-liang,et al.Offshore Wind Power Facilities Detection and Early Warning Assessment Based on Improved YOLOv5s[J].,2023,33(02):182.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 027]
[4]石露露,廖光忠.改进 YOLOv5s 的明渠漂浮垃圾实时检测方法[J].计算机技术与发展,2023,33(09):83.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 013]
　SHI Lu-lu,LIAO Guang-zhong.Real-time Detection Method of Floating Garbage in Open Channels Based on Improved YOLOv5s[J].,2023,33(02):83.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 013]
[5]魏雅丽,牛为华.改进 YOLOv5s 的轻量化航拍小目标检测算法[J].计算机技术与发展,2024,34(02):53.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 008]
　WEI Ya-li,NIU Wei-hua.Aerial Small Target Detection Based on Improved YOLOv5s Lightweight Algorithm[J].,2024,34(02):53.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 008]
[6]张元,关瑜,熊风光,等.基于改进Yolov5s的增强现实文物识别方法[J].计算机技术与发展,2024,34(07):17.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0118]
　ZHANG Yuan,GUAN Yu,XIONG Feng-guang,et al.An Augmented Reality Cultural Relics Recognition Method Based on Improved Yolov5s[J].,2024,34(02):17.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0118]
[7]陈钧,周井泉*,张志鹏.面向小目标的自校正YOLOv5检测增强算法[J].计算机技术与发展,2024,34(10):140.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0220]
　CHEN Jun,ZHOU Jing-quan*,ZHANG Zhi-peng.Self-calibrated YOLOv5 Detection Enhancement Algorithm for Small Targets[J].,2024,34(02):140.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0220]
[8]刘俊苗,王海晨.基于改进YOLOv5s的遥感图像车辆目标检测[J].计算机技术与发展,2024,34(12):87.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0283]
　LIU Jun-miao,WANG Hai-chen.Vehicle Target Detection in Remote Sensing Images Based on Improved YOLOv5s[J].,2024,34(02):87.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0283]

更新日期/Last Update: 2024-02-10