Safety Distance Monitoring Method for Personnel Live Working Based on Human Joint Point and Point Cloud Registration

WU Qinghua; GUO Jianlong; XIONG Shan; ZHOU Qingyun; XIA Shuang; WANG Lei

doi:10.13296/j.1001-1609.hva.2026.05.027

您当前的位置：

首页 >

文章列表页 >

Safety Distance Monitoring Method for Personnel Live Working Based on Human Joint Point and Point Cloud Registration

Technical Discussion | 更新时间：2026-05-13

- Safety Distance Monitoring Method for Personnel Live Working Based on Human Joint Point and Point Cloud Registration
- High Voltage Apparatus Vol. 62, Issue 5, Pages: 227-236(2026)
- 作者机构：
  
  广东电网有限责任公司培训与评价中心，广州 510520
  武汉新电电气股份有限公司，武汉 430073
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2026.05.027
  CLC：
- Received：26 November 2025，
  
  Revised：2026-01-10，
  
  Published：16 May 2026
- 稿件说明：
移动端阅览
WU Qinghua, GUO Jianlong, XIONG Shan, et al. Safety Distance Monitoring Method for Personnel Live Working Based on Human Joint Point and Point Cloud Registration[J]. High Voltage Apparatus, 2026, 62(5): 227-236.
DOI：

WU Qinghua, GUO Jianlong, XIONG Shan, et al. Safety Distance Monitoring Method for Personnel Live Working Based on Human Joint Point and Point Cloud Registration[J]. High Voltage Apparatus, 2026, 62(5): 227-236. DOI： 10.13296/j.1001-1609.hva.2026.05.027.

摘要

受电力作业场景环境复杂且作业人员易被遮挡等因素的影响，基于目标检测与定位的距离检测方法难以满足防触电安全距离监测的实际需求。因此，文中提出了一种基于人体关节点和点云配准的人员近电作业安全距离监测方法。该方法首先针对AlphaPose中两阶段目标检测方法速度慢、姿态参数确定效率低及泛化性差的问题，采用YOLOv8替换原有人员目标检测模块，并提出参数姿态非极大值抑制方法来进行姿态参数的快速确定，进而采用改进后的人体姿态估计网络对输入图像进行作业人员快速目标检测和人体姿态估计，得到作业人员目标检测框及关节点位置信息；然后将作业人员关节点数据与作业场景点云数据进行配准融合获取含有作业人员空间位置信息的点云；最后利用作业人员关节点与带电设备的空间位置信息计算安全距离，实现作业人员近电作业安全距离的实时监测与预警。测试结果表明，文中所提出的基于YOLOv8-s和改进AlphaPose的人体姿态估计方法的精度、速度分别达到87.03%、22.88 FPS。应用案例表明，文中方法对作业人员进行近电作业安全距离监测最大误差仅为88.70 mm且平均推理时间为61.20 ms，满足电力作业过程人员触电安全风险管控的实际要求。文中方法可广泛应用于电力作业或其他场景人员作业过程安全监测等领域。

Abstract

The distance detection mathods based on object detection and localization are hard to meet the practical requirements of anti-electric shock safety distance monitoring due to the factors sch as the complex environment of power operation scenarios and the fact that workers are easily onstructed.Therefore

a safety distance monitoring method for human anti-electric shock based on human joint point and point cloud registration is proposed in this paper. As for such issues as low speed

low efficiency of pose parameter determination and weak generalization of the two-stage object detection method in AlphaPose

in this method the VOLOv8 is adopted to replace the original human detection module

and a parametric pose non-maximum suppression methos is proposed to rapidly determine pose parameters. Furthermore

an improved human pose estimatino network is used to perform fast worker detection and human pose estimation in the input image

thereby obtaining the worker's detection bounding box and joint point position information. Then

the data of human joints and scene point cloud are fused to obtain the point cloud containing the spatial location information of workers. Finally

the safe distance is calculated by using the spatial location information of the human joints and the live equipment to achieve the real-time monitoring and early warning of operator's safe distance. The test results show that the accuracy and speed of the human pose estimation method proposed based on YOLOv8-s and improved AlphaPose are up to 87.03% and 22.88 FPS respectively.The application cases show that the method proposed in this paper achieves a maximum error of only 88.70 mm and an average inference time of 61.20 ms for monitoring the safe distance of personnelworking near electricity

meeting the practical requirements for managing the risk of electric shock to personnel during power operations.

关键词

Keywords

references

张远建,陈搏卿,姜炫丞,等.电力系统临近电安全距离预警系统研究[J].电气工程学报,2022,17(1):186-191.

ZHANG Yuanjian,CHEN Boqing,JIANG Xuancheng,et al. Research on early warning system of electric safety distance near electric power system[J]. Journal of Electrical Engineering,2022, 17(1):186-191.

邹德华,严宇,王伟,等.基于ToF技术的高压输电线路带电作业安全监测方法[J].高电压技术,2020,46(7):2570-2577.

ZOU Dehua,YAN Yu,WANG Wei,et al.Method of live working safety inspection for high voltage transmission lines based on ToF technology[J]. High Voltage Engineering,2020,46(7):2570-2577.

REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE, 2016:779-788.

马靖,任博文,陈来军,等.面向变电站近电作业的高精度多人姿态估计方法[J].浙江电力,2024,43(9):97-106.

MA Jing,REN Bowen,CHEN Laijun,et al. A high-precision multihuman body pose estimation approach for near-electricity work in substations[J]. Zhejiang Electric Power,2024,43(9):97-106.

卓浩泽,张驰,李泰霖,等.面向电力作业无人机的位姿估计与目标感知方法[J].南方电网技术,2025,19(7):140-149.

ZHUO Haoze,ZHANG Chi,LI Tailin,et al.Pose estimation and target perception methods for electric power operation unmanned aerial vehicles[J]. Southern Power System Technology,2025,19(7):140-149.

朱有产,郑怡,秦金磊.基于改进YOLOv3的绝缘子目标检测研究[J].电瓷避雷器,2022(3):166-171.

ZHU Youchan,ZHENG Yi,QIN Jinlei.Insulator target detection based on improved YOLOv3[J]. Insulators and Surge Arresters, 2022(3):166-171.

宋立业,刘帅,王凯,等.基于改进EfficientDet的电网元件及缺陷识别方法[J].电工技术学报,2022,37(9):2241-2251.

SONG Liye,LIU Shuai,WANG Kai,et al.Identification method of power grid components and defects based on improved EfficientDet[J]. Transactions of China Electrotechnical Society,2022,37(9):2241-2251.

杨乐,郭一鸣,霍勇博,等.改进YOLOv5在电力生产违规穿戴检测中的应用[J].电力系统保护与控制,2023,51(14):160-168.

YANG Le,GUO Yiming,HUO Yongbo,et al.Application ofimproved YOLOv5 for illegal wearing detection in electric power construction[J]. Power System Protection and Control,2023,51 (14):160-168.

CHARLES R Q,HAO Su,MO Kaichun,et al. Pointnet:Deep learning on point sets for 3D classification and segmentation[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway:IEEE,2017:77-85.

万京松,王城,本丽莉,等.面向电力场景的自动化作业流程关键技术研究[J].电子设计工程,2025,33(9):120-124.

WAN Jingsong,WANG Cheng,BEN Lili,et al.Research on key technologies of automated workflow for power scenarios[J]. Electronic Design Engineering,2025,33(9):120-124.

QI C R,LIU Wei,WU Chenxia,et al.Frustum PointNets for 3D object detection from RGB-D data[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2018:918-927.

WANG Zhixin,JIA Kui.Frustum convnet:Sliding frustums to aggregate local point-wise features for amodal 3D object detection[C]//2019IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Piscataway:IEEE,2019:1742-1749.

刘永刚,于丰宁,章新杰,等.基于激光点云与图像融合的3D目标检测研究[J].机械工程学报,2022,58(24):289-299.

LIU Yonggang,YU Fengning,ZHANG Xinjie,et al.Research on 3D object detection based on laser point cloud and image fusion[J]. Journal of Mechanical Engineering,2022,58(24):289-299.

徐斌,赵鹏,刘伟,等.融合图像与点云处理的托盘定位系统[J].制造业自动化,2022,44(8):157-161.

XU Bin,ZHAO Peng,LIU Wei,et al.Pallet positioning system integrating image and point cloud processing[J]. Manufacturing Automation,2022,44(8):157-161.

CHONG Tonghong,ZHANG Yanan,MA Chao,et al.Design and analysis of video desensitization algorithm based on lightweight model PP PicoDet[C]//2023 International Conference on Artificial Intelligence and Automation Control.Piscataway:IEEE,2023:121-124.

王雷雄,王波,马富齐,等.基于双目立体匹配和场景元素识别的变电人员近电安全距离检测方法研究[J].电网技术,2023, 47(3):1010-1021.

WANG Leixiong,WANG Bo,MA Fuqi,et al.Research on the detection method of the approach distance between substation workers and the live equipment based on binocular stereo matching and scene element recognition[J]. Power System Technology,2023, 47(3):1010-1021.

李敏霞.基于机器视觉的变电设备安全距离检测方法研究[D].北京:华北电力大学,2019.

LI Minxia.Research on safe distance detection method of substation equipment based on machine vision[D]. Beijing:North China Electric Power University,2019.

徐果.基于机器视觉的多视图变电设备安全距离监测方法[D].上海:上海电机学院,2022.

XU Guo. Safety Distance monitoring method of multi view substation equipment based on machine vision[D]. Shanghai :Shanghai Dianji University,2022.

王长海,陈倩,唐欣,等.基于点云与图像决策级融合的道路行人检测[J].广西大学学报(自然科学版),2021,46(6):1592-1601.

WANG Changhai,CHEN Qian,TANG Xin,et al. Road pedestrian detection based on point cloud and image decision-level fusion[J]. Journal of Guangxi University(Natural Science Edition),2021,46 (6):1592-1601.

张一睿,陶劲松,申政,等.输电线路杆塔周围人身安全风险评估与接地装置结构优化[J].智慧电力,2025,53(6):78-84.

ZHANG Yirui,TAO Jinsong,SHEN Zheng,et al.Assessment of personal safety risks around transmission line towers and structural optimization of grounding devices[J]. Smart Power ,2025,53(6):78-84.

乔嘉赓,曾智,王娴,等.改进YOLOv5s算法的配电网设备安全隐患辨识仿真[J].电子设计工程,2025,33(16):32-3541.

QIAO Jiageng,ZENG Zhi,WANG Xian,et al.Simulation of identifying safety hazards in distribution network equipment using improved YOLOv5s algorithm[J]. Electronic Design Engineering, 2025,33(16):32-3541.

冯可,翁进荣,吴川彬,等.电力检修安全作业无人机智能巡检YOLOv5目标检测[J].电子设计工程,2025,33(21):134-137.

FENG Ke,WENG Jinrong,WU Chuanbin,et al.Intelligent inspection of YOLOv5 target detection using UAV for power maintenance safety operations[J]. Electronic Design Engineering, 2025,33(21):134-137.

邱慧,虞悦,汤琳.基于YOLOv5和回溯算法的变电站二次接线图纸连接关系智能识别[J].电器与能效管理技术,2025 (11):59-64.

QIU Hui,YU Yue,TANG Lin.Intelligent recognition of connection relationships in secondary wiring drawings of substations based on YOLOv5 and backtracking algorithm[J]. Electrical & Energy Management Technology,2025(11):59-64.

金鑫,程凌森,赵亮,等.基于改进YOLOv5模型的智能变电站目标违规行为检测方法研究[J].电测与仪表,2024,61(8):179-185.

JIN Xin,CHENG Lingsen,ZHAO Liang,et al.Research on intelligent substation target violation detection method based on improved YOLOv5 model[J]. Electrical Measurement &Instrumentation,2024,61(8):179-185.

叶采萍,陈炯,马显龙,等.改进YOLOv5的变电站反无人机目标检测算法[J].南方电网技术,2024,18(2):89-97.

YE Caiping,CHEN Jiong,MA Xianlong,et al.Anti UAV target detection algorithm for substation based on improved YOLOv5[J]. Southern Power System Technology,2024,18(2):89-97.

肖俊阳,李远,苏适,等.基于改进YOLOv5s的电网异物检测算法[J].智慧电力,2024,52(7):16-23.

XIAO Junyang,LI Yua,SU Shi,et al.Grid foreign matter detection algorithm based on improved YOLOv5s[J]. Smart Power,2024,52 (7):16-23.

李欣,吕干云,龚彧,等.基于二维时频谱图与改进YOLOv5的电能质量扰动识别[J].浙江电力,2024,43(10):35-44.

LI Xin,LYU Ganyun,GONG Yu,et al.PQD recognition using twodimensional time - frequency spectrograms and an im - proved YOLOv5[J]. Zhejiang Electric Power,2024,43(10):35-44.

HAO Shufang,XIE Shuqin,YU Wingtai,et al.RMPE:Regional multi-person pose estimation[C]//2017 IEEE International Conference on Computer Vision (ICCV).Italy:IEEE,2017.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research on Temperature Measurement of Key Components in Transmission Lines Based on C2f-FADC-YOLOv8 and SE-ORB-VGG16

Live Working Insulation Blanket Surface Defect Recognition Based on Improved YOLOv8

Study on the Discharge Decomposition Mechanism of Environmentally-friendly Medium HFO-1336mzz（E）

Method for Determining the Explosion and Fire Location of High Voltage Bushings with Multi-region Attention Mechanism

Optimization Design for Thermal Conductivity of Graphene-modified Copper-based Electrical Contact Composite Materials

Related Author

WANG Yong

XU Yuncheng

WU Youpeng

TAN Fangxiong

ZHAO Yun

WU Tian

WANG Mingxin

TANG Pan

Related Institution

国网甘肃省电力公司酒泉供电公司，甘肃酒泉 735000State Grid Gansu Electric Power Company, Jiuquan Power Supply Company, Gansu Jiuquan 735000, China

College of Electrical Engineering and New Energy, China Three Gorges University

Hubei Provincial Engineering Technology Research Center for Power Transmission Line, China Three Gorges University

State Key Laboratory of Power Grid Environmental Protection, China Electric Power Research Institute

⁰