计算机与现代化

• 算法设计与分析 • 上一篇    下一篇

基于远程监控图像的绝缘子覆冰检测算法

  

  1. 1.国网江西省电力公司检修分公司,江西南昌330029;2.上海交通大学电子信息与电气工程学院,上海200240
  • 收稿日期:2014-04-04 出版日期:2014-05-28 发布日期:2014-05-30
  • 作者简介:钟幼平(1977-),男,江西上高人,国网江西省电力公司检修分公司高级工程师,博士,研究方向:多媒体处理,监控系统安全; 杨威(1967-),男,浙江杭州人,工程师,学士,研究方向:电力系统架构,电力监控; 周洋(1991-),男,江苏南京人,上海交通大学电子信息与电气工程学院硕士研究生,研究方向:计算机视觉; 林巍峣(1980-),男,浙江瑞安人,副教授,博士,研究方向:多媒体处理,视频监控。
  • 基金资助:
    国家自然科学基金资助项目(61001146); 上海市浦江人才计划项目(12PJ1404300); 国网江西省电力公司检修分公司项目

An Image-based Algorithm for Icing Detection of Insulator

  1. 1. Maintenance Branch, National Grid Jiangxi Electric Power Co., Nanchang 330029, China;

    2. School of Electronic Information and Electrical Engineering, Shanghai Jiaotong University, Shanghai 200240, China
  • Received:2014-04-04 Online:2014-05-28 Published:2014-05-30

摘要: 提出一种基于远程监控图像的绝缘子覆冰情况检测算法。本文算法充分利用边缘检测方法和其他图像处理方法,对覆冰绝缘子侧面冰棱的边缘情况进行精确的检测。本文算法还引入了摄像头的标定参数,使其能更加准确地估计覆冰绝缘子侧面的冰棱长度。当摄像头因为外力或其他因素而产生角度偏移时,本文算法能及时地对摄像头标定参数做出修正,并利用新的标定参数重新估计绝缘子的覆冰情况。实验结果表明,本文算法对绝缘子侧面冰棱边缘检测效果显著,满足实时监控和预警报警的要求。

关键词: 覆冰检测, 绝缘子, 监控图像, 厚度估计, 相机标定

Abstract:  In this paper, an algorithm based on edge detection for icing detection of insulator was proposed. It took full advantage of the edge detection algorithm and other image processing method to make more specific and accurate detection of icing insulator. Also, it introduced the camera calibration to estimate the length and thickness of swing ice on the side of insulators. What’s more, when the settled angle of camera shifted due to some outside force or other reason, this algorithm was able to revise the camera calibration parameters in time and use them to estimate the length and thickness of ice again. The experimental results show that the detection of icing insulator edge is pretty good and the estimation of the length and thickness of ice is accurate, which definitely meet the requirement of real-time monitoring and early warnings.

Key words:  icing detection, insulator, surveillance image, thickness estimation, camera calibration

中图分类号: