[1]赵成伟,吴云东,蔡国榕,等.基于超体素的LiDAR点云粘连目标分割算法[J].集美大学学报(自然科学版),2017,22(1):73-80.
ZHAO Chengwei,WU Yundong,CAI Guorong,et al.Research of Segmentation Algorithm for LiDAR Point Cloud Based on Supervoxel[J].Journal of Jimei University,2017,22(1):73-80.
点击复制
基于超体素的LiDAR点云粘连目标分割算法(PDF)
《集美大学学报(自然科学版)》[ISSN:1007-7405/CN:35-1186/N]
- 卷:
-
第22卷
- 期数:
-
2017年第1期
- 页码:
-
73-80
- 栏目:
-
数理科学与信息工程
- 出版日期:
-
2017-01-28
文章信息/Info
- Title:
-
Research of Segmentation Algorithm for LiDAR Point Cloud Based on Supervoxel
- 作者:
-
赵成伟1; 2; 吴云东3; 蔡国榕3; 陈水利4
-
(1.集美大学理学院,福建 厦门 361021;2.厦门市无人机遥感应用工程技术研究中心,福建 厦门 361021;3.集美大学计算机工程学院, 福建 厦门 361021;4.集美大学诚毅学院, 福建 厦门 361021)
- Author(s):
-
ZHAO Chengwei1; 2; WU Yundong3; CAI Guorong3; CHEN Shuili4
-
(1.School of Science,Jimei University,Xiamen 361021,China;2.Xiamen VAVRS Application Engineering TechnologyResearch Center,Xiamen 361021,China;3.Computer Engineering College,Jimei University,Xiamen 361021,China;4.Chengyi University College,Jimei University,Xiamen 361021,China)
-
- 关键词:
-
超体素; 归一化分割; 点云; 分割算法
- Keywords:
-
supervoxel; graph cut; point cloud; segmentation algorithm
- 分类号:
-
-
- DOI:
-
-
- 文献标志码:
-
A
- 摘要:
-
针对点云地物分割结果中存在的粘连现象,结合三维点云的空间分布和颜色信息,引入过分割方法将点云集划分为超体素,并构建加权图模型。在此基础上,利用归一化分割方法实现点云粘连区域的目标分割。针对树木、建筑物的实验结果表明,该方法对树木之间、树木与建筑物之间的粘连具有良好的分类效果。
- Abstract:
-
Due to the existence of adhesion phenomenon in the point cloud segmentation,by using combined with the spatial distribution and color information of the three- dimensional point cloud,this paper proposes an over-segmentation method to divide the point clouds into supervoxels,and then gives a weighted graph model.Hence,we realize the segmentation of point clouds adhesion areas using graph cut.In view of the trees,buildings,the experiment results show that the method has a good classification effect about the adhesion between the trees and buildings.
参考文献/References:
-
相似文献/References:
更新日期/Last Update:
2017-03-09