光学学报, 2017, 37 (10): 1011001, 网络出版: 2018-09-07
激光点云的混合流形谱聚类自适应分割方法 下载: 717次
Mixed Manifold Spectral Clustering Adaptive Segmentation Method for Laser Point Cloud
图 & 表
图 1. 仿真点集的分割结果。(a)两条曲线构成的螺旋线;(b)相交的直线与平面
Fig. 1. Segmentation results of simulated point sets. (a) Spirals made by two curves; (b) crossed lines and plane
图 2. 符合目标点云几何结构的分割结果。(a) Rabbit; (b) Horse; (c) Armadillo
Fig. 2. Segmentations of point clouds based on target geometry. (a) Rabbit; (b) Horse; (c) Armadillo
图 5. 点云分割结果一致性与kNN之间的关系
Fig. 5. Relationship between consistency of segmentation and value of kNN
图 6. 不同分割算法对含噪声的点云图像的分割结果。(a)添加噪声的点云;(b) k-means算法;(c)局部角距离谱聚类算法;(d)本文算法
Fig. 6. Segmentation results of point clouds with noises by different segmentation algorithms. (a) Point cloud with noise; (b) k-means algorithm; (c) local angular distance spectral clustering algorithm; (d) proposed algorithm
图 7. 卫星模型点云的分割结果。(a)灰度图像;(b)切片式三维成像法得到的侧视图;(c)本文算法得到的正视图; (d)本文算法得到的侧视图
Fig. 7. Segmentation results of point cloud of satellite model. (a) Gray image; (b) side view of slice three-dimensional imaging; (c) front view of segmentation by the proposed algorithm; (d) side view of segmentation by the proposed algorithm
王帅, 孙华燕, 郭惠超, 都琳. 激光点云的混合流形谱聚类自适应分割方法[J]. 光学学报, 2017, 37(10): 1011001. Shuai Wang, Huayan Sun, Huichao Guo, Lin Du. Mixed Manifold Spectral Clustering Adaptive Segmentation Method for Laser Point Cloud[J]. Acta Optica Sinica, 2017, 37(10): 1011001.