基于YOLO的无人机电力线路杆塔巡检图像实时检测

doi:10.11930/j.issn.1004-9649.201812028

中国电力 ›› 2019, Vol. 52 ›› Issue (7): 17-23.DOI: 10.11930/j.issn.1004-9649.201812028

• 泛在电力物联网——先进信息与通信技术 • 上一篇下一篇

基于YOLO的无人机电力线路杆塔巡检图像实时检测

郭敬东¹, 陈彬¹, 王仁书¹, 王佳宇², 仲林林²

1. 国网福建省电力有限公司电力科学研究院, 福建福州 350007;
2. 东南大学电气工程学院, 江苏南京 210096

收稿日期:2018-12-12 修回日期:2019-03-19 出版日期:2019-07-05 发布日期:2019-07-13
作者简介:郭敬东(1968-),男,高级工程师,从事信息通信、信息系统开发、信息自动化、智能信息处理研究,E-mail:guo_jingdong@fj.sgcc.com.cn;仲林林(1990-),男,通信作者,博士,讲师,从事高电压与放电等离子体基础理论与应用、大数据与人工智能技术研究,E-mail:linlin@seu.edu.cn
基金资助:
国家自然科学基金资助项目（518070280）；国家电网有限公司科技项目（52130418000L）。

YOLO-Based Real-Time Detection of Power Line Poles from Unmanned Aerial Vehicle Inspection Vision

GUO Jingdong¹, CHEN Bin¹, WANG Renshu¹, WANG Jiayu², ZHONG Linlin²

1. State Grid Fujian Electric Power Research Institute, Fuzhou 350007, China;
2. Department of Electrical Engineering, Southeast University, Nanjing 210096, China

Received:2018-12-12 Revised:2019-03-19 Online:2019-07-05 Published:2019-07-13
Supported by:
This work is supported by National Natural Science Foundation of China (No.51807028), and Science and Technology Project of SGCC (No.52130418000L).

摘要/Abstract

摘要： 无人机巡检已成为电力线路灾后巡检的重要方式。然而，目前的无人机巡检仍主要通过人工方式评估线路灾损，不仅费时费力，而且准确率低。提出了一种基于深度学习算法（YOLO）的实时目标检测模型，用于灾后根据无人机巡检视频实时检测电力杆塔的状态。通过对倒断类杆塔图像进行数据增广，解决了杆塔类别不平衡问题。通过使用K-means算法对杆塔数据集的目标框进行重新聚类，改进了YOLO算法参数。测试结果表明，该模型能有效检测多种环境下多种尺度的杆塔目标。改进后的模型在测试集上的召回率和交并比（IoU）较改进前有所提高，且平均均值精度（mAP）达到94.09%，检测速度达到20帧/s。此外，也对更快的简化版YOLO模型进行了测试，检测速度能达到30帧/s。

关键词: 无人机巡检, 电力杆塔, 深度学习, YOLO, 数据增广, 人工智能与大数据应用

Abstract: Unmanned aerial vehicles (UAV)-based inspection has become an important approach for power line inspection after disaster. However, the current UAV-based inspection is still performed manually for damage assessments, which is not only time-consuming but also poor in accuracy. In this paper a real-time detection model based on YOLO deep learning algorithm is presented to detect the status of power line poles automatically from the UAV vision data after disaster. The data augmentation is performed for collapsed towers to solve the class imbalance problem. To improve the parameters of YOLO, K-means algorithm is used to cluster object frames of pole data. The experimental results show that the proposed model can effectively detect multi-scale towers in multiple environments. The Recall and Intersection-over-Union (IoU) of the improved YOLO are improved, with the mean average precision (mAP) on the test set of 94.09% and the average processing speed of 20 frames per second (FPS) after improving the parameters. Moreover, we tested the simplified YOLO with faster speed, and the average processing speed reaches 30 FPS.

Key words: UAV inspection, power line poles, deep learning, YOLO, data augmentation, artificial intelligence and big data application

中图分类号:

TM755

郭敬东, 陈彬, 王仁书, 王佳宇, 仲林林. 基于YOLO的无人机电力线路杆塔巡检图像实时检测[J]. 中国电力, 2019, 52(7): 17-23.

GUO Jingdong, CHEN Bin, WANG Renshu, WANG Jiayu, ZHONG Linlin. YOLO-Based Real-Time Detection of Power Line Poles from Unmanned Aerial Vehicle Inspection Vision[J]. Electric Power, 2019, 52(7): 17-23.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201812028

https://www.electricpower.com.cn/CN/Y2019/V52/I7/17

参考文献

[1] 韩冰, 尚方. 面向无人机输电线路巡检的电力杆塔检测框架模型[J]. 浙江电力, 2016, 35(4):6-11 HAN Bing, SHANG Fang. Power tower detection framework model for UAV transmission line inspection[J]. Zhejiang Electric Power, 2016, 35(4):6-11
[2] KATRASNIK, PERNUS, LIKAR. A survey of mobile robots for distribution power line inspection[J]. IEEE Transactions on Power Delivery, 2009, 25(1):485-493.
[3] ROVERSO D, NGUYEN V N, JENSSEN R. Automatic autonomous vision-based power line inspection:a review of current status and the potential role of deep learning[J]. International Journal of Electrical Power & Energy Systems, 2018, 99(7):107-120.
[4] TOTH J, GILPIN-JACKSON A. Smart view for a smart grid:unmanned aerial vehicles for transmission lines[C]//International Conference on Applied Robotics for the Power Industry Montreal. Montreal, Canada:IEEE, 2010:1-6.
[5] DENG C, WANG S, HUANG Z, et al. Unmanned aerial vehicles for power line inspection:a cooperative way in platforms and communications[J]. Journal of Communications, 2014, 9(9):687-692.
[6] 章绍耕. 浅谈"抗台风"电网抢修物资供应过程中所应注意的问题[J]. 农电管理, 2017(3):28-29 ZHANG Shaogeng. A brief discussion on the problems in the material supply and emergency repair process of power grid in anti-typhoon[J]. Rural Power Management, 2017(3):28-29
[7] 王万国, 田兵, 刘越, 等. 基于RCNN的无人机巡检图像电力小部件识别研究[J]. 地球信息科学学报, 2017, 19(2):256-263 WANG Wanguo, TIAN Bing, LIU Yue, et al. Research on power component recognition of unmanned aerial vehicle inspection image based on RCNN[J]. Geo-Information Science, 2017, 19(2):256-263
[8] 杨晓旭, 温招洋. 深度学习在输电线路绝缘子故障检测中的研究与应用[J]. 中国新通信, 2018, 20(10):208-210 YANG Xiaoxu, WEN Zhaoyang. Research and application of deep learning in fault detection of transmission line insulators[J]. China New Telecommunications, 2018, 20(10):208-210
[9] NORDENG I E, HASAN A, OLSEN D, et al. DEBC detection with deep learning[C]//Scandinavian Conference on Image Analysis. Troms, Norway:Springer, 2017:248−259.
[10] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149.
[11] LIU W, ANGUELOV D, ERHAN D, et al. SSD:single shot multibox detector[C]//European Conference on Computer Vision. Amsterdam, Netherlands:Springer, 2016:21-37.
[12] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified, real-time object detection[C]//Computer Vision and Pattern Recognition. Las Vegas, USA:IEEE, 2016:779-788.
[13] REDMON J, FARHADI A. YOLO9000:better, faster, stronger[C]//IEEE Conference on Computer Vision and Pattern Recognition. Hawaii, USA:IEEE, 2017:6517−6525.
[14] REDMON J, FARHADI A. YOLOv3:an incremental improvement[DB/OL]. (2018-04-08)[2018-10-18]. http//arXiv.org/pdf/1804.02767.pdf.
[15] HE K, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]//Conference on Computer Vision and Pattern Recognition. Hawaii, USA:IEEE, 2017:2980-2988.
[16] MARTINEZ C, SAMPEDRO, CHAUHAN A, et al. Towards autonomous detection and tracking of electric towers for aerial power line inspection[C]//International Conference on Unmanned Aircraft Aystems. Orlando, USA:IEEE, 2014:284−295.
[17] 苑津莎, 崔克彬, 李宝树. 基于ASIFT算法的绝缘子视频图像的识别与定位[J]. 电测与仪表, 2015, 52(7):106-112 YUAN Jinsha, CUI Kebin, LI Baoshu. Recognition and location of insulator video and image by ASIFT-based algorithm[J]. Electrical Measurement Instrumentation, 2015, 52(7):106-112
[18] GERKE M, SEIBOLD P. Visual inspection of power lines by UAS[C]//International Conference and Exposition on Electrical and Power Engineering. Iasi, Romania:IEEE, 2014:284−295.
[19] KRAWCZYK B. Learning from imbalanced data:open challenges and future directions[J]. Progress in Artificial Intelligence, 2016, 5(4):1-12.
[20] OUYANG W, WANG X, ZHANG C, et al. Factors in fine tuning deep model for object detection with long-tail distribution[C]//Computer Vision and Pattern Recognition. Las Vegas, USA:IEEE, 2016:864−873.
[21] ENGSTROM L, TSIPRAS D, SCHMIDT L, et al. A rotation and a translation suffice:fooling CNNs with simple transformations[DB/OL]. (2017-12-07)[2018-01-20]. http//arXiv.org/pdf/1712.02779.pdf.

基于YOLO的无人机电力线路杆塔巡检图像实时检测

YOLO-Based Real-Time Detection of Power Line Poles from Unmanned Aerial Vehicle Inspection Vision

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	姜通海, 王峰, 刘子琪, 单帅杰. 基于改进生成对抗网络的风光气象资源联合场景生成方法[J]. 中国电力, 2025, 58(3): 183-192.
[2]	宋智伟, 黄新波, 纪超, 张凡, 张烨. 基于PCSA-YOLOv7 Former的输电线路连接金具及其锈蚀检测方法[J]. 中国电力, 2024, 57(6): 141-152.
[3]	刘东, 张越, 皮俊波, 单连飞, 刘赫, 宋鹏程, 姜涛. 面向电网调控信息智能检索的知识图谱构建及应用[J]. 中国电力, 2023, 56(7): 78-84.
[4]	张鑫, 叶俊杰, 崔瑶, 黄鑫, 仲林林. 基于语义信息距离解耦的变电运维多类别缺陷图像检测[J]. 中国电力, 2023, 56(6): 209-218.
[5]	周震震, 宋云海, 何宇浩, 王黎伟, 黄和燕, 何珏, 朱志航, 闫云凤. 基于分组查询注意力的可扩展电力人员行为分类方法[J]. 中国电力, 2023, 56(11): 77-85.
[6]	李冰, 白云山, 赵宽, 郭聪彬, 翟永杰. 基于HSCA-YOLOv7的风电机组叶片表面缺陷检测算法[J]. 中国电力, 2023, 56(10): 43-52.
[7]	李丰君, 王磊, 赵健, 张建宾, 张世尧, 田杨阳. 基于天气融合和LSTM网络的分布式光伏短期功率预测方法[J]. 中国电力, 2022, 55(11): 149-154.
[8]	赵永良, 付鑫, 郭阳, 边迎迎, 王思宁. 基于深度学习和图像识别的电力配件智能出入库[J]. 中国电力, 2021, 54(3): 55-60.
[9]	马静怡, 崔昊杨, 张明达, 孙益辉, 许永鹏. 基于改进Faster RCNN的小尺度入侵目标识别及定位[J]. 中国电力, 2021, 54(3): 38-44.
[10]	黄郑, 王红星, 周航, 张星炜, 赵宏伟. 基于混合算法的电力杆塔巡检实时航迹规划[J]. 中国电力, 2021, 54(11): 214-220.
[11]	周俊煌, 黄廷城, 谢小瑜, 范纹郡, 易婷婷, 张勇军. 视频图像智能识别技术在输变电系统中的应用研究综述[J]. 中国电力, 2021, 54(1): 124-134,166.
[12]	吴云亮, 张建新, 李豹, 李鹏, 李智勇, 周鑫, 杨燕, 赖晓文. 深度学习辅助约束辨识的电力市场快速出清方法[J]. 中国电力, 2020, 53(9): 90-97,207.
[13]	赵会茹, 赵一航, 郭森. 基于互补集合经验模态分解和长短期记忆神经网络的短期电力负荷预测[J]. 中国电力, 2020, 53(6): 48-55.
[14]	杨晓楠, 孙博, 郎燕生. 基于深度学习的特高压直流闭锁故障智能调度决策[J]. 中国电力, 2020, 53(6): 8-17.
[15]	张焕坤, 李军毅, 张斌. 基于改进型YOLO v3的绝缘子异物检测方法[J]. 中国电力, 2020, 53(2): 49-55.

模态框（Modal）标题

基于YOLO的无人机电力线路杆塔巡检图像实时检测

YOLO-Based Real-Time Detection of Power Line Poles from Unmanned Aerial Vehicle Inspection Vision

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics