This paper proposes a three-step approach to extract power lines from components with mobile laser scanning (MLS) data. First, we use the maximum a posteriori estimation to partition point clouds into components. Each component contains points from one object only. The segmentation is optimized by the minimum-cost perfect matching globally and robust to Gaussian noise with the help of the proposed robust estimator. Then, we extract power lines from components based on the linear structure information. Finally, power line components are grouped into individual spans. Experiments show that our method succeeds to achieve the power line extraction from MLS data effectively and outperforms state-of-the-art approaches in terms of the accuracy and robustness.
Recommended citation: ‘S. Xu, R. Wang(*). "Power Line Extraction From Mobile LiDAR Point Clouds." IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. vol. 12(2), pp. 734-743, 2019.’