利用LiDAR数据进行电力设施提取与建模可以克服传统工程测量电力巡线工作量大,危险性高,效率低下等缺点,但现有的电力线提取研究主要集中在电力线的分离与提取,并且拟合的精度不高。针对此问题本文提出了一种精度较高的电力线拟合方法。首先,根据电力线两端悬挂、中间自然下垂的特点,求解电力线拟合的最佳几何模型;然后,通过电力线的走向和端点,建立电力线拟合的最佳平面坐标系;最后,采用基于二次多项式限制的最小二乘法拟合电力线,解算出最优参数,生成最终电力线模型。对真实数据的处理和精度评价表明,本文方法不仅能够实现电力线的快速3维重建,而且能够达到较高的拟合精度。

High-voltage power transmission line is an important infrastructure of a country, the breakdown of power facilities may bring huge damage to the daily lives of the people and the economy of the country. Thus, management and monitoring of power facilities is important. Traditional engineering measures have the drawbacks of high workload, high risk, and low efficiency. Airborne LiDAR can overcome these drawbacks, and cannot be easily affected by environment, making LiDAR an important development trend for power line measurement. At present, airborne LiDAR is mainly used for the reconstruction of urban areas and natural features; extraction and modeling of power facilities is still in its infancy. The main problems are too much human intervention, low accuracy, and lack of continuity. Therefore, this paper proposes a new method based on the analysis of the characteristics of power facilities and extraction algorithm. First, the optimal catenary fitting geometry model of the powerline was obtained based on the powerline characteristics. The nonlinear catenary equation was simplified to linear polynomial form based on the principle of Lagrange polynomial to improve the operation efficiency. Second, the projection direction of the powerline in the XOY plane was determined based on a statistic method. Afterward, the vertical plane that contains the starting and end points and are perpendicular to the power line were also determined based on the characteristic of the end laser point. The distance of all points to the plane as X-axis parameters was calculated and the points corresponding to the Z coordinate values as Y-axis parameters were obtained to determine the optimal plane coordinate system. Finally, the quadratic-polynomial-limited least square method was used to fit the powerlines, obtain the optimal parameters, and generate the final power line model. Experiments on real data showed that the proposed method improved the facility and veracity of powerline fitting. The precision reached ±1.740 cm both in the vertical and horizontal planes. The proposed method can quickly fit the powerline and reach a high accuracy, which are significant in the study of 3D reconstruction by LiDAR point cloud.