Point Cloud Registration Method for Power Transmission Lines based on Combined Feature Points

In view of the problems that traditional point cloud registration algorithms are inefficient and prone to local optimality when processing large-scale power transmission line point cloud data, this paper proposes a point cloud registration method based on combined feature points (CFMPCR). First, the adaptive octree is used to downsample the point cloud, and then the combined feature points are extracted by combining 3D-SIFT and Alpha Shape algorithms, and the feature points are matched by FPFH. The SAC-IA algorithm is used for coarse registration, and the improved ICP algorithm is used for fine registration, that is, the iterative closest point algorithm from point to surface combined with KD-Tree for acceleration. Experimental results show that CFMPCR is significantly superior to the traditional ICP registration algorithm in both registration accuracy and efficiency. The registration accuracy is improved by about two orders of magnitude, and the registration time is reduced by 80%. The method CFMPCR proposed in this paper has good robustness and registration accuracy when processing large-scale point cloud data, and the registration effect is better than the existing algorithm.