系统仿真学报 ›› 2016, Vol. 28 ›› Issue (9): 2154-2158.

• 短文 • 上一篇    下一篇

KD树索引策略下紧支撑径向基函数的点云建模

李佳1,2,3,4, 段平1, 盛业华2,3,4, 吕海洋2,3,4, 张思阳2,3,4   

  1. 1.云南师范大学旅游与地理科学学院,云南 昆明 650500;
    2.虚拟地理环境教育部重点实验室(南京师范大学),江苏 南京 210023;
    3.江苏省地理环境演化国家重点实验室培育建设点,江苏 南京 210023;
    4.江苏省地理信息资源开发与利用协同创新中心,江苏 南京 210023
  • 收稿日期:2015-05-28 修回日期:2015-07-30 出版日期:2016-09-08 发布日期:2020-08-14
  • 作者简介:李佳(1984-),女,湖北公安,博士,讲师,研究方向为图形图像处理;段平(通讯作者1984-),男,湖北监利,博士,讲师,研究方向为GIS三维建模。
  • 基金资助:
    国家自然科学基金(41271383)

Point Cloud Modeling Based on Compactly Supported Radial Basis Function under KD Tree Index Strategy

Li Jia1,2,3,4, Duan Ping1, Sheng Yehua2,3,4, Lü Haiyang2,3,4, Zhang Siyang2,3,4   

  1. 1. School of Tourism and Geographical Sciences of Yunnan Normal University, Kunming 650500, China;
    2. Key Laboratory of Virtual Geographic Environment (Nanjing Normal University), Ministry of Education, Nanjing 210023, China;
    3. State Key Laboratory Cultivation Base of Geographical Environment Evolution (Jiangsu Province), Nanjing 210023, China;
    4. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, China
  • Received:2015-05-28 Revised:2015-07-30 Online:2016-09-08 Published:2020-08-14

摘要: 采用紧支撑径向基函数(Compactly Supported Radial Basis Function,CSRBF)对点云进行建模和可视化表达会因穷举搜索问题导致计算机内存溢出,最终引起建模和可视化失败。KD树索引具有快速搜索点的优点且避免了穷举搜索问题,将KD树索引和CSRBF插值模型相结合,提出KD树搜索策略下的CSRBF点云建模与表达方法。建立点云数据的KD树索引,采用CSRBF构建点云的隐式曲面函数模型,通过Marching Cubes算法对建好的模型进行有效的可视化表达。采用经典的兔子点云进行实验验证,结果表明KD树索引搜索策略下CSRBF的点云建模与表达方案可行。

关键词: 点云, 建模, KD树, 紧支撑径向基函数

Abstract: Modeling constructed by Compactly Supported Radial Basis Function (CSRBF) and visualization will fail. The main reason is that exhaustive search results in out of memory. KD tree can void exhaustive search due to the advantage of quickly searching. CSRBF combined KD tree was used to construct point cloud model. Modeling approach of CSRBF based on KD tree was proposed. KD tree index of the point cloud was constructed. CSRBF interpolation method was used to construct implicit function model of point cloud. Marching Cubes algorithm was to display model in 3D manner. The experimental results with rabbit point cloud show that it is feasible for point cloud modeling with CSRBF based on KD tree.

Key words: point cloud, modeling, KD tree, compactly supported radial basis function

中图分类号: