基于局部几何特征的点云分割算法

o deal with the segmentation accuracy losing issue caused by over-segmentation in the existing 3D point cloud segmentation algorithms, a novel geometric features based point cloud segmentation algorithm is proposed in this paper. A local weighted curvature feature calculation method together with a triangular mesh-based concave and convex feature point selection algorithm are proposed to get candidate feature point set of segmentation. The candidate feature point sets are selected and their intersections are selected respectively. Final local geometricsegmentation point set are then collected by threshold and weighted calculation, which resolved the edge information loss issue caused by over-segmentation in the traditional algorithms. The consensus cutting algorithm is applied as the last step to split the point cloud. Experimental results show that the proposed algorithm improves the segmentation accuracy for the complex 3D point cloud model.