摘要
为进一步减轻输电线路进行定期检查、巡视的任务,文章提出了一种利用智能化无人机巡检技术,对航拍图像进行线路的提取和跟踪。采用直方图均衡化及图像滤波对航拍图像进行预处理,解决了航拍图像光照强度以及背景对输电线路元素提取的干扰;采用LSD算法实现了线路边缘的提取,在去除图像背景信号的基础上使用Hough变换数学算法实现了输电线路的准确连接;分别采用粒子滤波和扩展卡尔曼滤波两种图像跟踪方法对航拍视频进行线路跟踪,通过建立输电线运动模型,用仿真软件对其进行识别,两种方法的检测准确度分别为95.34%和94.72%,证实文章处理算法可实现输电线路的提取和跟踪。
In order to further alleviate the tasks of regular inspections and patrols of transmission lines,the article proposes an intelligent drone inspection technology to extract and track aerial images.Firstly,the aerial image is preprocessed by using histogram equalization and image filtering to solve the interference of the aerial image light intensity and background on the extraction of transmission line elements;secondly,the LSD algorithm is used to achieve the extraction of the line edge,and the image background signal is removed.On the basis of the Hough transform mathematical algorithm,the accurate connection of the transmission line is realized;finally,the particle filter and the extended Kalman filter are used to track the aerial video line by the two image tracking methods.To identify them,the detection accuracy of the two methods are 95.34% and 94.72%,respectively,confirming that the article processing algorithm can realize the extraction and tracking of transmission lines.
作者
沙伟燕
何宁辉
丁培
胡伟
李秀广
周秀
Sha Weiyan;He Ninghui;Ding Pei;Hu Wei;Li Xiuguang;Zhou Xiu(Electric Power Research Institute of State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan 750011,China;State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan 750001,China)
出处
《电测与仪表》
北大核心
2022年第5期158-165,共8页
Electrical Measurement & Instrumentation
基金
宁夏自然科学基金资助项目(2018AAC03222)。
