一种基于机载激光点云的门型电塔精确提取方法

A method for accurate extraction of gated electric towers based on airborne laser point cloud

随着无人机技术的发展,越来越多的电力部门采用机载激光雷达及倾斜摄影测量技术代替传统的人工方法,进行电力巡检、电力部件检修等工作。门型电塔为目前电力走廊中比较常见的一种高压电塔。在无数据标签的电网点云数据中,对门型电塔进行精确点云提取是门型电塔日常维护的重要工作。本文从包含有地面、植被、电力线、门型电塔的电网点云数据入手,首先对点云进行体素化分割,对体素网格进行高程连续与高程阈值的筛选,并对电塔顶部进行扩展搜索,达到去除地面点及植被点的目的。然后根据门型电塔水平投影成条状的特点,利用门型电塔的圆柱杆塔部分得到门型电塔的水平投影直线,再通过距离阈值判断去除点云中多余的电力线点。本文介绍的方法实现了从无标签的电网点云数据中对门型电塔的精确提取,为进一步进行门型电塔的三维建模、维护检修等工作提供了数据基础。

With the development of unmanned earial vehicle(UAV)technology,more and more power departments use airborne light detection and ranging(LiDAR)technology and oblique photogrammetry technology to replace the traditional manual method gradually,and carry out power inspection,power component maintenance and other work.As a common high-voltage tower in the power corridor at present,accurate point cloud extraction is an important step in the daily maintenance of the gated electric tower in the unlabeled grid point cloud data.In this paper,starting with the grid point cloud data including the ground,vegetation,power line and portal tower,the point cloud is divided into voxel and voxel grid.In order to remove the ground point and vegetation point,the line elevation is continuous and the elevation threshold is filtered,and the top of the tower is extended to search.Then the horizontal projection straight line of the gated electric tower is obtained by using the characteristics of the horizontal projection of the gated electric tower,and the redundant power line point cloud is removed by using the distance threshold judgment.The method introduced in this paper realizes the accurate extraction of the gated electric tower from the unlabeled grid point cloud data,and provides data basis for further three-dimensional modeling,maintenance and repair of the gated electric tower.