赵 博,马宏忠,张 潇,李春亮,赵金雄,张学军,张 琴.定向识别航拍绝缘子及其缺陷检测方法研究[J].电子测量与仪器学报,2023,37(5):240-251
定向识别航拍绝缘子及其缺陷检测方法研究
Research on directional identification of aerial insulators andtheir defect detection methods
  
DOI:
中文关键词:  绝缘子  定向识别  注意力机制  YOLOv5
英文关键词:insulator  targeted identification  attention mechanisms  YOLOv5
基金项目:甘肃省教育厅产业支撑项目(2022CYZC-38)、国网甘肃省电力公司科技项目(522722220013)、甘肃省自然基金(21JR7RA282)项目资助
作者单位
赵 博 1. 国网甘肃省电力公司 
马宏忠 2. 国网甘肃省电力公司电力科学研究院 
张 潇 3. 兰州交通大学电子与信息工程学院 
李春亮 1. 国网甘肃省电力公司 
赵金雄 2. 国网甘肃省电力公司电力科学研究院 
张学军 3. 兰州交通大学电子与信息工程学院 
张 琴 4. 国网兰州供电公司 
AuthorInstitution
Zhao Bo 1. State Grid Gansu Electric Power Company 
Ma Hongzhong 2. State Grid Gansu Electric Power Research Institute 
Zhang Xiao 3. School of Electronic and Information Engineering, Lanzhou Jiaotong University 
Li Chunliang 1. State Grid Gansu Electric Power Company 
Zhao Jinxiong 2. State Grid Gansu Electric Power Research Institute 
Zhang Xuejun 3. School of Electronic and Information Engineering, Lanzhou Jiaotong University 
Zhang Qin 4. State Grid Lanzhou Power Supply Company 
摘要点击次数: 988
全文下载次数: 870
中文摘要:
      针对现有绝缘子检测算法无法定向检测绝缘子及其缺陷的问题,提出了一种改进 YOLOv5( you only look once v5, YOLOv5)算法的航拍绝缘子识别及其缺陷检测方法。 通过定向标注航拍绝缘子图片,形成航拍绝缘子数据集和缺陷绝缘子数 据集;在 YOLOv5 的主干特征提取网络引入轻量化注意力机制模块、在特征融合阶段使用改进的空间金字塔池化结构;通过改 进 YOLOv5 网络的头部结构使其可以对绝缘子进行定向识别,并对损失函数添加角度损失分类。 实验结果表明在检测时间由 单张 0. 044 s 到单张 0. 049 s 并无显著增长的前提下,改进后的算法在测试集上的 mAP(mean average precision)的值为 95. 00%, 实现了定向识别绝缘子及其漏帽缺陷,还可应用到绝缘子视频流检测。 为后续的绝缘子精确定位以及进一步故障检测打下良 好基础。
英文摘要:
      Aiming at the problem that the existing insulator detection algorithm cannot detect insulators and their defects in an oriented manner, an aerial insulator identification and defect detection method improved by YOLOv5 algorithm is proposed. By orienting the aerial insulator pictures, the aerial insulator dataset and defective insulator dataset are formed. The lightweight attention mechanism module is introduced in the backbone feature extraction network of YOLOv5, and the improved spatial pyramid pooling structure is used in the feature fusion stage. By improving the head structure of the YOLOv5 network, the network can perform directional identification of insulators and add angular loss classification to the loss function. The experimental results show that under the premise that the detection time does not increase significantly from 0. 044 s to 0. 049 s per sheet, the value of mAP (mean average precision) on the test set of the improved algorithm is 95. 00%, which realizes directional identification of insulators and their leakage cap defects, and can also be applied to insulator video stream detection. This provides a good basis for the subsequent precise positioning of insulators and further fault detection.
查看全文  查看/发表评论  下载PDF阅读器