陈志忠,熊泽森,姚东,郑欢,宋维铜,杨志新,贾涛
CHEN Zhizhong,XIONG Zesen,YAO Dong,ZHENG Huan,SONG Weitong,YANG Zhixin,JIA Tao

• 1:广东电网有限责任公司汕尾供电局
• 2:武汉大学遥感信息工程学院

摘要(Abstract)：

电力杆塔主要构件的图像识别是无人机巡检的主要内容，准确识别杆塔构件对保障电网运行具有重要价值。为此，提出一种基于深度学习和知识图谱的电力杆塔主要构件识别方法。首先，建立不同构件类型的拓扑关系，形成杆塔空间知识图谱；其次，设计语义关系推理模型，融合构件语义特征与拓扑关系，得到增强特征；最后，拼接增强特征与原始特征，实现特征融合。实验表明：在未架线电力杆塔多目标识别方面，所提方法比Reasoning-RCNN、Cascade-RCNN及Faster-RCNN的识别效果好，能够精准识别杆塔主要构件，对无人机电力巡检具有参考价值。
The image recognition of the main components of electric power towers is a primary focus of UAV inspections, as accurately identifying these tower components holds significant value for ensuring the smooth operation of power grids. To address this need, the paper proposes a method for recognizing the main components of electric power towers based on deep learning and knowledge graph. Firstly, the paper establishes topological relationships between component types, forming a spatial knowledge graph of the towers. Subsequently, it designs a model for semantic relationship inference that integrates semantic features of components with their topological relationships, resulting in feature enhancement. Finally, by concatenating these enhanced features with the original features, feature fusion is achieved. Experimental results demonstrate that the proposed method outperforms Reasoning-RCNN, Cascade-RCNN, and Faster-RCNN in the multi-target recognition of unstrung towers. It enables precise recognition of the main tower components, thus offering valuable insights for UAV-based power line inspections.

关键词(KeyWords)： 深度学习;电力杆塔;智能识别;知识图谱;Reasoning-RCNN
deep learning;electric power tower;intelligent recognition;knowledge graph;Reasoning-RCNN

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金资助项目（41971332）;; 南方电网科技项目（0315002022030201JJ00025）

作者(Author): 陈志忠,熊泽森,姚东,郑欢,宋维铜,杨志新,贾涛
CHEN Zhizhong,XIONG Zesen,YAO Dong,ZHENG Huan,SONG Weitong,YANG Zhixin,JIA Tao

DOI: 10.19585/j.zjdl.202405012

参考文献(References)：

• [1]顾晓东，唐丹宏，黄晓华.基于深度学习的电网巡检图像缺陷检测与识别[J].电力系统保护与控制，2021,49(5):91-97.GU Xiaodong,TANG Danhong,HUANG Xiaohua.Deep learning-based defect detection and recognition of a power grid inspection image[J]. Power System Protection and Control,2021,49(5):91-97.
• [2]陈亦平，赵曼勇，刘文涛，等.南方电网连锁故障风险分析及防御措施研究[J].南方电网技术，2010,4(3):8-11.CHEN Yiping,ZHAO Manyong,LIU Wentao,et al.The risks analysis and defensive measures research of cascading failures in China southern power grid[J]. Southern Power System Technology,2010,4(3):8-11.
• [3]甘德强，胡江溢，韩祯祥.2003年国际若干停电事故思考[J].电力系统自动化，2004,28(3):1-4.GAN Deqiang,HU Jiangyi,HAN Zhenxiang. After the2003 blackouts across several continentals[J].Automation of Electric Power Systems,2004,28(3):1-4.
• [4] CHEN H Y,WANG X J,LI Z H,et al.Distributed sensing and cooperative estimation/detection of ubiquitous power Internet of Things[J]. Protection and Control of Modern Power Systems,2019,4(2):151-158.
• [5]余贻鑫.智能电网实施的紧迫性和长期性[J].电力系统保护与控制，2019,47(17):1-5.YU Yixin. Urgency and long-term nature of smart grid implementation[J].Power System Protection and Control,2019,47(17):1-5.
• [6]陈维江，赵国亮.以新能源为主体的新型电力系统关键技术和装备[J].全球能源互联网，2022,5(1):1.CHEN Weijiang,ZHAO Guoliang. Key technologies and equipment of new power system with new energy as the main body[J].Global Energy Internet,2022,5(1):1.
• [7]刘志颖，缪希仁，陈静，等.电力架空线路巡检可见光图像智能处理研究综述[J].电网技术，2020,44(3):1057-1069.LIU Zhiying,MIAO Xiren,CHEN Jing,et al.Review of visible image intelligent processing for transmission line inspection[J]. Power System Technology,2020,44(3):1057-1069.
• [8]柳长安，叶文，吴华等.融合地理位置信息的电力杆塔检测[J].华中科技大学学报（自然科学版），2013,41（增刊1）:208-210.LIU Changan,YE Wen,WU Hua,et al. Detection of power poles based on fusion of geographical location information[J].Journal of Huazhong University of Science and Technology(Natural Science Edition),2013,41(S1):208-210.
• [9]王孝余，李丹丹，张立颖.一种基于监督学习的输电线监测中杆塔的检测方法[J].东北电力技术，2017,38(11):12-14.WANG Xiaoyu,LI Dandan,ZHANG Liying. A supervised approach to electric tower detection for power line inspection[J].Northeast Electric Power Technology,2017,38(11):12-14.
• [10]韦汶妍，刘晓立，杨传凯，等.一种基于深度学习网络的输电杆塔智能检测算法[J].广东电力，2020,33(9):166-173.WEI Wenyan,LIU Xiaoli,YANG Chuankai,et al.An intelligent detection algorithm for transmission tower based on deep learning network[J].Guangdong Electric Power,2020,33(9):166-173.
• [11]李晓峰，胡川，尹洪，等.基于机载LiDAR点云的电力线自动提取[J].计算机测量与控制，2018,26(7):220-224.LI Xiaofeng,HU Chuan,YIN Hong,et al.Power line extraction based on airborne LiDAR[J].Computer Measurement&Control,2018,26(7):220-224.
• [12]骆顾平，陈静，苏成悦，等.复杂地物背景下的电力线识别算法[J].应用科技，2021,48(5):23-28.LUO Guping,CHEN Jing,SU Chengyue,et al.A power line recognition algorithm under the background of complex ground objects[J].Applied Science and Technology,2021,48(5):23-28.
• [13]张从新，赵乐，王先培.复杂地物背景下电力线的快速提取算法[J].武汉大学学报（工学版），2018,51(8):732-739.ZHANG Congxin,ZHAO Le,WANG Xianpei.Research on fast extraction algorithm of power line in complex ground object background[J].Engineering Journal of Wuhan University,2018,51(8):732-739.
• [14]张俊，庞世强，李晓斌，等.融合LSD算法与Hough变换的航拍输电线路图像杆塔自动识别方法[J].电子器件，2021,44(5):1210-1214.ZHANG Jun,PANG Shiqiang,LI Xiaobin,et al. Automatic tower recognition method based on LSD algorithm and Hough transform in aerial transmission line image[J].Chinese Journal of Electron Devices,2021,44(5):1210-1214.
• [15]司小庆，束庆霏.基于电力巡检的图像识别研究[J].电力大数据，2022,25(10):37-44.SI Xiaoqing,SHU Qingfei.Research of image recognition based on electric power inspection[J].Power Systems and Big Data,2022,25(10):37-44.
• [16] GIRSHICK R,DONAHUE J,DARRELL T,et al.Region-based convolutional networks for accurate object detection and segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(1):142-158.
• [17] REN S,HE K,GIRSHICK R,et al. Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE Trans Pattern Anal Mach Intell,2017,39(6):1137-1149.
• [18]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. Study on the electrical devices detection in UAV images based on region based convolutional neural networks[J].Journal of Geo-Information Science,2017,19(2):256-263.
• [19]罗潇，於锋，彭勇.基于深度学习的无人机电网巡检缺陷检测研究[J].电力系统保护与控制，2022,50(10):132-139.LUO Xiao,YU Feng,PENG Yong.UAV power grid inspection defect detection based on deep learning[J].Power System Protection and Control,2022,50(10):132-139.
• [20]张蕾.基于R-CNN的绝缘子图像定位研究[D].北京：华北电力大学，2018.ZHANG Lei.Research on the localization of insulator images based on R-CNN[D].Beijing:North China Electric Power University,2018.
• [21]杨焰飞，曹阳.基于深度学习的无人机拍摄图像绝缘子目标检测[J].激光杂志，2020,41(10):63-66.YANG Yanfei,CAO Yang.Image insulator target detection based on deep learning for UAV[J]. Laser Journal,2020,41(10):63-66.
• [22]史晋涛，李喆，顾超越，等.基于样本扩充的Faster RCNN电网异物监测技术[J].电网技术，2020,44(1):44-51.SHI Jintao,LI Zhe,GU Chaoyue,et al.Research on foreign matter monitoring of power grid with faster R-CNN based on sample expansion[J]. Power System Technology,2020,44(1):44-51.
• [23] XU H,JIANG C H,LIANG X D,et al. ReasoningRCNN:Unifying adaptive global reasoning into largescale object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Long Beach,CA,USA:IEEE,2019.
• [24] CAI Z W,VASCONCELOS N. Cascade R-CNN:high quality object detection and instance segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,43(5):1483-1498.