Advanced Search

ISSN1001-3806 CN51-1125/TN Map

Volume 46 Issue 1
Jan.  2022
Article Contents
Turn off MathJax

Citation:

Power line extraction from airborne LiDAR data based on cloth simulation

  • Received Date: 2020-08-20
    Accepted Date: 2020-10-19
  • In order to realize the power line extraction of long distance linear airborne light detection and ranging(LiDAR), a power line extraction method of long-distance airborne LiDAR based on cloth simulation was proposed. On the basis of data preprocessing, the function between cloth and corresponding airborne LiDAR point cloud was analyzed by simulating the falling process of cloth. The position where cloth stayed after gravity falling was determined as the power line point cloud with similar height, and then the straight line was fitted in the xOy plane. The distances from points to the fitting line were used to judge the number of power lines whether odd or even. And the power line point clouds were divided by the judgment of points on both sides of the line to achieve the extraction of a single power line. The experimental results show that the accuracy of the proposed method is 98.9%, with high degree of automation and not sensitive to the lack of local point cloud, which has good engineering application value for intelligent powerline inspection and automatic analysis of transmission channel spatial structure.
  • 加载中
  • [1]

    CHEN C, YANG B S, SONG S, et al. Automatic clearance anomaly detection for transmission line corridors utilizing UAV-borne LiDAR data[J]. Remote Sensing, 2018, 10(4): 613-633. doi: 10.3390/rs10040613
    [2]

    WU H, LIU H Y, DING G F, et al. Automatic extraction of power lines from laser point clouds in complex environments[J]. Laser Technology, 2020, 44(4): 509-514(in Chinese).
    [3]

    SHI L, GUO T, PENG Ch, et al. Segmentation of laser point cloud and safety detection of power lines[J]. Laser Technology, 2019, 43(3): 341-346(in Chinese).
    [4]

    WANG Y J, CHEN Q, LIU L, et al. A hierarchical unsupervised method for power line classification from airborne LiDAR data[J]. International Journal of Digital Earth, 2019, 12(12): 1406-1422. doi: 10.1080/17538947.2018.1503740
    [5]

    SHI H Y, GUO T, WANG D, et al. Power line suspension point location method based on laser point cloud[J]. Laser Technology, 2020, 44(3): 364-370(in Chinese).
    [6]

    PENG X Y, SONG S, QIAN J J, et al. Research on automatic positioning algorithm of power transmission towers based on UAV LiDAR[J]. Power System Technology, 2017, 41(11): 3670-3677(in Chin-ese).
    [7]

    CHEN Ch, PENG X Y, SONG Sh, et al. Safety distance diagnosis of large scale transmission line corridor inspection based on LiDAR point cloud collected with UAV[J]. Power System Technology, 2017, 41(8): 2723-2730(in Chinese).
    [8]

    YU J, MU Ch, FENG Y M, et al. Power lines extraction techniques from airborne LiDAR data[J]. Geomatics and Information Science of Wuhan University, 2011, 36(11): 1275-1279 (in Chinese).
    [9]

    CHEN C, MAI X M, SONG S, et al. Automatic power lines extraction method from airborne LiDAR point cloud[J]. Geomatics and Information Science of Wuhan University, 2015, 40(12): 1600-1605(in Chinese).
    [10]

    MANOHAR Y, CHARUDATTA G. Extraction of power lines using mobile LiDAR data of roadway environment[J]. Remote Sensing Applications: Society and Environment, 2017, 8: 258-265. doi: 10.1016/j.rsase.2017.10.007
    [11]

    ZHAO L, WANG X P, DAI D D, et al. Automatic extraction algorithm of power line in complex background[J]. High Voltage Engineering, 2019, 45(1): 218-227(in Chinese).
    [12]

    WU J J, CHEN L, LI L, et al. Power line extraction and reconstruction from airborne LiDAR point cloud[J]. Laser Technology, 2019, 43(4): 500-505(in Chinese).
    [13]

    ZHANG C X, ZHAO L, WANG X P, et al. Fast extraction algorithm of power lines in complex ground objects[J]. Journal of Wuhan University(Engineering Edition), 2018, 51(8) : 732-739(in Chin-ese).
    [14]

    WANG J, XIA Sh B, WANG H P, et al. Study on reconstruction of bundled conductors from helicopter-borne LiDAR data[J]. Remote Sensing Technology and Application, 2015, 30(6): 1189-1194(in Chinese).
    [15]

    LIN X G, ZHANG J X. 3D power line reconstruction from airborne LiDAR point cloud of overhead electric power transmission corridors[J]. Acta Geodatrica et Cartographica Sinica, 2016, 45(3): 347-353 (in Chinese).
    [16]

    DUAN M Y. 3D power line reconstruction from airborne LiDAR point cloud[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(12): 1495(in Chinese).
    [17]

    LAI X D, DAI D C, ZHENG M, et al. Power line 3D reconstruction form LiDAR point cloud data[J]. Journal of Remote Sensing, 2014, 18(6): 1223-1229(in Chinese).
    [18]

    McLAUGHLIN R A. Extracting transmission lines from airborne LiDAR data[J]. IEEE Geoscience & Remote Sensing Letters, 2006, 3(2): 222-226.
    [19]

    SOHN G, JWA Y, KIM H B. Automatic powerline scene classification and reconstruction using airborne LiDAR data[C]// ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences 2012. Melbourne, Australia: International Society for Photogrammetry and Remote Sensing, 2012: 167-172.
    [20]

    WEIL J. The synthesis of cloth objects[J]. ACM SIGGRAPH Computer Graphics, 1986, 20(4): 49-54. doi: 10.1145/15886.15891
    [21]

    ZHANG W M, QI J B, WAN P, et al. An easy-to-use airborne LiDAR data filtering method based on cloth simulation[J]. Remote Sensing, 2016, 8(6): 501-519. doi: 10.3390/rs8060501
    [22]

    WANG G, WANG Q, LIU Sh T, et al. Method of building extraction from UAV oblique photography point cloud based on cloth simulation[J]. Bulletin of Surveying and Mapping, 2020(10): 97-100.
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(7) / Tables(1)

Article views(5156) PDF downloads(22) Cited by()

Proportional views

Power line extraction from airborne LiDAR data based on cloth simulation

  • 1. Institute of Civil Engineering, Henan University of Engineering, Zhengzhou 451191, China
  • 2. School of Geography and Environmental Science, Tianjin Normal University, Tianjin 300387, China
  • 3. School of Electrical and Information Engineering, Hunan University, Changsha 410082, China

Abstract: In order to realize the power line extraction of long distance linear airborne light detection and ranging(LiDAR), a power line extraction method of long-distance airborne LiDAR based on cloth simulation was proposed. On the basis of data preprocessing, the function between cloth and corresponding airborne LiDAR point cloud was analyzed by simulating the falling process of cloth. The position where cloth stayed after gravity falling was determined as the power line point cloud with similar height, and then the straight line was fitted in the xOy plane. The distances from points to the fitting line were used to judge the number of power lines whether odd or even. And the power line point clouds were divided by the judgment of points on both sides of the line to achieve the extraction of a single power line. The experimental results show that the accuracy of the proposed method is 98.9%, with high degree of automation and not sensitive to the lack of local point cloud, which has good engineering application value for intelligent powerline inspection and automatic analysis of transmission channel spatial structure.

引言
  • 智能电网已逐渐成为中国电网行业发展的新方向。随着智能电网的持续推进,输电廊道的空间结构进行高精度、自动化分析需求不断增加,作为电力系统稳定运行的重要保障,输电线路的3维数据获取及安全监测具有重要的意义。机载激光雷达能够全天候作业,具备高精度高效率获取电网3维信息的优势,能快速获取电力线3维坐标,克服传统摄影测量因影像分辨率等原因而无法进行电力线测量的缺憾,在电力巡线工作方面具有重要的现实意义[1]。然而,电力线周边环境复杂,周围生长的树木以及电力设施周边的建(构)筑物等都会对电力的安全运行构成潜在威胁[2-3], 因此,如何从获取的激光雷达3维数据中高效、自动地进行电力线提取,对电力走廊设备监测及设备之间的空间关系分析十分重要[4-7]

    针对激光雷达电力线信息的自动提取问题,国内外学者开展了大量的研究,常用的方法包含两大类:一是基于2维图像处理领域中的Hough变换进行电力线检测[8-12],该类方法利用较为成熟的图像处理知识,具备操作简单的特点,但对于垂直排列的多根电力线情况,该方法无法检测,此外,处理过程中将激光雷达点云退化为栅格,会对原始点云数据带来精度损失;另一类利用3维点云空间直线进行电力线检测[13-17],该类方法将电力线分档进行处理,一方面依赖分档参数,缺乏长距离电力线整体的考虑,另一方面易受激光雷达点云数据量和地形起伏等因素的影响。除此之外,有学者采用监督分类的方法进行电力先分离,如McLAUGHLIN尝试通过局部仿射模型从激光雷达数据中进行电力线提取[18],该方法依赖初始模型,初始模型的选择对提取精度产生直接影响。SOHN等人试图利用马尔可夫随机场模型进行电力线和建筑物分离[19],该方法需大量训练样本,样本的选择会对结果产生影响,同时不均匀采样也会导致错分。

    由于激光雷达(light detection and ranging, LiDAR)点云存在不规则性和易产生局部数据缺失等特点, 以及真实3维场景的复杂性,从机载LiDAR数据中提取电力线仍是一个值得深入研究的难题。作者受布料模拟滤波思想[20]的启发,提出一种基于布料模拟的长距离机载激光雷达电力线提取方法。

1.   长距离机载激光雷达电力线提取方法
  • 作为典型的人工构造物,电力线3维点云数据在空间上具有狭长分布、相近高度通常具有多根电力线、单根电力线之间相互平行的特点,本文中提出的方法主要包括点云数据预处理、布料模拟法相近高度电力线识别、直线拟合与单根电力线点提取等步骤。

  • 在电力线走廊区域,机载激光点云通常包括地面、植被、电塔和电力线。本文中首先利用基于布料模拟的滤波方法[21]滤除地面点,保留非地面点云,根据先验知识,电力线高度距地面不小于3.0m,设定距离地面高度阈值为3.0m,非地面点云中按照距离地面布料网格高度小于3.0m的点进一步滤除,不小于3.0m的点保留,供进一步提取电力线。

  • 布料模拟最早由WEIL提出的3维悬挂式布料模拟建模方法[20],与以往基于点云数据本身进行识别的方法不同,布料模拟方法从外部的布料入手,将布料下落的过程进行计算机模拟,分析布料下落过程中布料与点云之间的作用[22],从而可以把电力线点云作为一个整体,具有参量设置少和电力线跨档整体提取的特点。

    质点弹簧模型通常被用于布料模拟建模,如图 1所示。

    Figure 1.  Schematic diagram of mass spring model

    质点弹簧模型中,布料通过粒子构建网格进行建模,需要计算出所有粒子的位置,根据牛顿第二运动定律:

    式中,m为粒子质量,X(t)为粒子在时间t的位置,Fext(X, t)为粒子运动方向上作用在粒子上的外力,Fint(X, t)为粒子间相互作用产生的内力,Xt表示产生粒子间内力位置和时间。由(1)式可知,作用在粒子上的力决定了粒子的位置及速度。

    布料模拟算法应用于同一高度附近的电力线提取中,粒子的运动只需考虑并限制在高程方向上,将粒子在高程方向与电力线点比较,进行碰撞检测,如果粒子高度与电力线点高度一致,粒子停止移动,位置只由重力决定,即(1)式中的粒子间相互作用的内力为零,对(1)式进行求解得:

    式中,Δt为时间步长,G为常量。如果Δt和粒子初始位置已知,则粒子当前位置可通过(2)式计算得到。

    粒子会在不同电力线间的空隙发生移动,为避免此种粒子移动发生,粒子在重力下移动后的内力需考虑,此时,粒子将在网格中并回到初始位置。由于粒子的移动方向被限制,因此, 粒子之间相互连接的弹簧两端不同高度的粒子会移动到同一水平面。如果弹簧两端的粒子都能移动,将两端的粒子沿相反方向移动相同的位移量,若弹簧一端的粒子不可移动,则移动另外一端的粒子。若弹簧两端的粒子高度相同,则粒子不发生移动。粒子移动的位移量按下式计算:

    式中,d为粒子运动的位移量; b为标示符,取值为0和1,代表粒子可移动性; p0是粒子移动前的位置; pi为粒子最终位置; n为竖直方向的矢量。粒子的重复运动,可通过对参量刚度进行设置用于描述粒子重复运动的次数。整个过程如下:设置布料分辨率,将布料粒子和激光雷达点云进行水平投影,在投影面内找到布料粒子最近的激光雷达点作为粒子的对应点,用P表示,其投影前的高程为HP,代表布料能下降到的最低高度。记粒子移动中的高度为H,通过迭代计算HHP,如果HHP,则将粒子移到布料能下降到的最低高度HP,并设置粒子可移动性为不可移动。通过计算可得激光点云与粒子间的高度差,如果小于设定的阈值hth,则点云为识别出的同一高度附近的电力线点,并从待处理电力线中移除,作为进一步相近高度单个电力线提取的数据源,此过程不断迭代,直到达到设定电力线的不同高度数。

  • 根据先验知识,在同一个高度附近,通常存在两条以上电力线,并且不同的电力线点投影在xOy平面后存在平行关系,经过布料模拟算法处理之后,同一高度附近的电力线条数存在以下两种情况:(1)奇数条,如图 2所示;(2)偶数条,如图 3所示。

    Figure 2.  Schematic diagram of odd power line points at similar height

    Figure 3.  Schematic diagram of even power line points with similar height

    假定同一高度附近存在的电力线点数为n,在xOy平面内用最小二乘法拟合可得直线方程:

    式中,a1, a2, a3为直线方程的系数。分别计算所有电力线点到(4)式中直线的距离di(i=1, 2…, n),如下式所示:

    设定距离阈值dth,如果第i个点到拟合直线的距离di < dth,则该点位于拟合的直线上;设定单根电力线包含的最少点数为Pmin,如果位于直线上的点数大于Pmin,则表明该同一高度附近的电力线条数为奇数,同时将位于直线上的点从n个同一高度附近的电力线点中移除,并将其标记为1#电力线点,剩下的点则满足第2种情况,即同一高度附近的电力线条数为偶数。

    同一高度附近的电力线条数为偶数时,利用点是否分布与拟合直线的两侧作为判断条件,将同一高度附近的电力线点一分为二,并分别在xOy平面内进一步拟合直线,此过程不断迭代,直到参与拟合直线的点全部位于一条直线上为止,即剩下的所有点均在拟合的直线上,通过以上处理,单根完整电力线会被提取出来。

2.   实验
  • 选用某地的机载激光点云数据,该数据包含8971003个点,共有3个电塔跨越4档,电力线总长度为985.29m,宽度约145.4m,实验区地势起伏,最大高差为103.41m,包含植被、树木、电塔、电力线等地物,总共有5根架空输电线,受地理环境、作业天候、载体以及遮挡等多重因素影响,电力线点云数据存在多处局部数据缺失现象,主要集中在上部的两根电力线中,如图 4所示,放大区域为其中一处数据缺失情况。

    Figure 4.  Point cloud data of experimental area

    经过点云数据预处理,设定电力线的高度数为2,布料网格尺寸为2m,单次布料模拟迭代次数为500,LiDAR点到布料的距离阈值hth设为0.5m,得到两个不同的近似高度电力线点分别如图 5图 6所示。其中, 图 5包含2条电力线,图 6包含3条电力线。

    Figure 5.  Two power lines with height 1 after cloth simulation

    Figure 6.  Three power lines with height 2 after cloth simulation

    对于两个高度的电力线点,设置点到直线的阈值dth=0.2m,单根电力线上的最少点数Pmin=100,进行单根电力线提取,效果如图 7所示,不同的电力线点分别用不同的颜色显示。

    Figure 7.  Extracted power line data

  • 图 5图 6可以看出,经过布料模拟能够将不同近似高度的单根电力线成功提取,图 7及其放大部分的提取结果表明,经过本文中的方法提取的电力线点完整,且对电力线缺失部分不敏感。

    为进一步定量的分析提取效果,利用TerraScan软件对电力线点云手工分类,统计提取的点数,并与本文中的方法提取效果进行对比,如表 1所示。

    power line No. extraction results of theproposed method manually extraction results extraction rate/%
    1# 2608 2621 99.5
    2# 2601 2614 99.5
    3# 5032 5042 99.8
    4# 5827 5834 99.9
    5# 5207 5212 99.9
    total 21095 21323 98.9

    Table 1.  Accuracy analysis of the proposed method

    表 1可以看出,1#电力线和2#电力线由于存在较多局部点缺失,点数明显少于3#、4#、5#电力线点,进一步验证了本文中的方法能够实现长距离电力线的提取,且对局部点云缺失具有良好的抗干扰能力。分析主要原因如下:(1)利用原始的点云数据,避免数据内插带来的精度和信息损失;(2)通过布料模拟从电力线点的整体入手,避免了电力线点局部邻域尺度的选择和局部点缺失对提取结果的影响;(3)利用了输电线路相近高度存在多条电力线,且电力线近似相互平行的先验知识;(4)使用了点云滤波、直线拟合等成熟的算法和开源代码,提高了电力线提取的整体稳定性。

3.   结论
  • 针对长距离直线型机载激光雷达电力线提取,提出基于布料模拟的长距离机载激光点云电力线提取方法,将布料模拟算法引入到电力线点云提取中,该方法仅依靠原始点云数据,需要电力线高度数、布料网格尺寸、单次布料模拟迭代次数,LiDAR点到布料的距离阈值hth、点到直线的距离阈值dth和单根电力线最少点数Pmin等参量,将长距离电力线作为一个整体,无需考虑电力线的档距、分段参量。实验结果表明,本文中的方法能够实现长距离电力线的提取,对局部电力线点云缺失具有良好的抗干扰能力,且参量简单、自动化程度高,提供一种新的利用机载激光点云数据进行电力线提取方法。

Reference (22)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return