基于YOLOv5的图像检测技术在线路巡检中的应用  

Application of YOLOv5-based Image Detection Technology in the Inspection of Transmission Lines

作　　者：闫付乐 蔡永挚 段进丽 罗伟 郑琛琛 YAN Fule;CAI Yongzhi;DUAN Jinli;LUO Wei;ZHENG Chenchen(State Grid Beijing Electric Power Co.,Ltd.,Beijing 100051)

机构地区：[1]国网北京市电力公司,北京100051

出　　处：《电力安全技术》2024年第5期57-60,共4页Electric Safety Technology

摘　　要：小型化巡检机器人的应用可降低人工线路巡检的劳动强度并提高检测效率,同时改善人工巡检盲区等问题。介绍了小型化巡检机器人系统,同时基于YOLOv5及其改进算法,根据巡检机器人拍摄的图像制作数据集,并进行目标图像检测试验验证,表明改进后的算法具备更加精准的识别能力,有效保障了输电线路中设备的安全稳定运行。Smart inspection robots are applied to reduce the labor intensity in manual inspection of transmission lines,enhance the inspection efficiency,and improve the manual inspection of blind spots.A smart inspection robot system is introduced,and a data set is created based on YOLOv5 and its improved algorithm as well as the images captured by the inspection robots.Target image detection experiments conducted for verification show that the improved algorithm is of more accurate recognition capacity,which effectively ensures the safe and stable operation of equipment in transmission lines.

关 键 词：输电线路 图像检测 巡检机器人 YOLOv5算法 

分 类 号：TM75[电气工程—电力系统及自动化]