基于Hough变换和总体最小二乘法的电力线检测  被引量:18

Power Line Detection Based on Hough Transform and Total Least Squares Method

在线阅读下载全文

作  者:操昊鹏 曾卫明[1] 石玉虎[1] 徐鹏[1] CAO Hao-peng;ZENG Wei-ming;SHI Yu-hu;XU Peng(School of Information Engineering,Shanghai Maritime University,Shanghai 201306,China)

机构地区:[1]上海海事大学信息工程学院,上海201306

出  处:《计算机技术与发展》2018年第10期164-167,共4页Computer Technology and Development

基  金:国家自然科学基金(31470954)

摘  要:随着智能电网技术的飞速发展,无人机智能巡检已经受到了广泛关注,而电力线是电力故障检测和无人机导航的重要参照物。针对电力系统中故障检测定位难的问题,提出了一种基于霍夫变换和总体最小二乘法的电力线提取方法。首先,在原图像经过预处理的基础上,通过霍夫变换进行初步检测,获得电力线的大致位置;然后,根据霍夫变换的检测结果锁定电力线的范围;最后,对锁定范围内所有的点,使用总体最小二乘法精确拟合出电力线。实验结果表明,该方法能够很好地检测电力线位置并拟合出电力线,具有鲁棒性强、检测精度高等特点。该方法不仅适用于地形复杂的山地区域,而且对于图像的清晰度要求并不高,因而在电力系统故障检测中具有很强的实用性。With the rapid development of smart grid technology,UAV intelligent inspection has attracted a lot of people’s attention,andthe power line is an important reference for power fault detection and unmanned aerial navigation. Aimed at the problem of fault detectionand positioning in power system,we propose a power line extraction method based on Hough transform and total least squares method. Indetail,the Hough transform is firstly used to obtain the approximate position of the power line. Then,the range of the power line isroughly determined according to the detection result of the Hough transform. Finally,we use the overall least squares method for all thepoints in the locking range above which can fit the location of power line accurately. The experiment shows that the method proposed candetect the position of power line and fit the power line well,with strong robustness and high detection accuracy. The method is not onlyapplicable to mountainous areas with complex topography,but also has not high requirement for image definition,so it has strong practicability in power system fault detection.

关 键 词:智能电网 电力线 霍夫变换 总体最小二乘法 图像处理 

分 类 号:TP301[自动化与计算机技术—计算机系统结构]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象