To improve the effectiveness and generality of existing methods of high voltage (HV) transmission lines extraction, this paper proposes a novel automatic extraction method of HV transmission lines using airborne LIDAR point cloud data by incorporating the geography of transmission corridors and the distribution characteristics of airborne LIDAR point cloud data. The proposed method results in the separation of ground objects by using a differentiation height threshold segmentation algorithm based on subspace features, which divide long-distance space into several small-distance space sets to improve the generality of the algorithm. A height density segmentation algorithm is used to locate transmission towers and extract point cloud data of HV transmission lines to improve the efficiency of the algorithm. Feasibility test cases show that the proposed method can automatically extract HV transmission lines from airborne LIDAR point cloud data in flat terrains, canyon terrains, and steep slope terrains in an efficient manner, with high accuracy and generality.