Power line inspection is an important task for the daily maintenance and safety assurance of overhead transmission lines. Traditional manual inspection methods are time-consuming and labor-intensive, making it difficult to meet the growing demand for power line inspection. With the rapid development of laser scanning technology, airborne LiDAR plays a significant role in transmission line safety exploration and power line inspection by emitting laser pulses toward overhead transmission lines, enabling rapid acquisition of extensive three-dimensional spatial information of power lines. Based on this, this paper proposes a method for extracting power lines and high-voltage transmission towers from transmission line point cloud data. First, a quadric surface fitting-based point cloud filtering method is applied to eliminate a large number of invalid points. According to the spatial dimension characteristics of the point cloud, the original data is projected onto a horizontal plane and classified. Rough extraction of power lines is achieved by analyzing the elevation continuity of the point cloud. Subsequently, normalized elevation filtering is employed to effectively remove ground noise points and obtain accurate power line seed points. For non-ground points, downsampling is performed, and spatial dimension features are estimated based on the covariance matrix properties constructed from neighborhood point sets to acquire high-voltage transmission tower seed points. Finally, using the power line and transmission tower seed points, the kd-tree algorithm is utilized for nearest neighbor searching to obtain more complete power lines and high-voltage transmission towers. Experimental results demonstrate that this method can extract complete power lines and high-voltage transmission towers from laser point cloud data, exhibiting certain engineering value.