三维激光扫描仪在电力铁塔结构提取中的应用

Application of Terrestrial Laser Scanner in the Structure Extraction of Power Tower

针对目前电力铁塔结构提取与塔身倾斜度检测中存在方法单一、可靠性低等问题,提出了利用地面三维激光扫描仪对电力铁塔非接触式扫描,通过建立铁塔塔身独立坐标系、粗差剔除、点云建模与精度分析,准确提取出铁塔结构并计算塔身倾斜度,定量验证了铁塔处于安全运营状态。采用随机抽样一致性(RANSAC)算法进行原始点云粗差剔除,利用铁塔自身结构几何特征有效地确定了RANSAC算法中筛选阈值、有效数据比例、迭代次数等参数,实现了海量点云中铁塔结构点云数据的精确提取,弥补了传统最小二乘法、特征值法中参数选取的不足。

Aiming at the problems ot single method and low reliability in the structure extraction and inclination de- tection ot electric tower, this paper proposes a non-contact scanning ot electric tower by using Terrestrial Laser Scanner. By establishing the independent coordinate system ot the tower, gross error elimination, point cloud modeling and preci- sion analysis, the structure and inclination ot the tower are acquired exactly, which has veritied quantitatively that the tow- er is in a sate operating state. In this paper, the Random Sample Consensus （RANSAC） algorithm is used to eliminate the gross error in the original point cloud ot the tower, the parameters ot the RANSAC algorithm, such as the screening threshold,the effective data ratio and the number ot iterations, are effectively determined by the geometrical teatures ot the tower itselt, which achieves extraction ot the tower structure point cloud data fiom huge amount ot the original point cloud data, making up the traditional least squares method and the eigenvalue method for the selection ot parameters.