标记点过程用于点云建筑物提取

徐文学; 杨必胜; 董震; 彭向阳; 麦晓明; 王珂; 高文武

doi:10.13203/j.whugis20130044

标记点过程用于点云建筑物提取

¹ 武汉大学测绘遥感信息工程国家重点实验室,湖北武汉,430079;² 国家海洋局第一海洋研究所,山东青岛,266061;³ 广东电力科学研究院,广东广州,510080

基金项目: 国家973计划资助项目 (2012CB725301)；国家自然科学基金资助项目 (41071268)；教育部博士点基金资助项目(20120141110035)；南方电网公司重点科技资助项目(K-GD2013-030)

详细信息

作者简介:
徐文学,博士。主要从事激光扫描点云数据处理与三维重建研究。

通讯作者:
高文武

中图分类号: P237.4
计量
- 文章访问数: 03823
- HTML全文浏览量: 068
- PDF下载量: 0991
出版历程
- 收稿日期: 2013-04-14
- 修回日期: 2014-05-04
- 发布日期: 2014-05-04

Building Extraction from Point Cloud Using Marked Point Process

¹State Key Laboratory of Information Engineering in Surveying,Mapping,and Remote Sensing,Wuhan University,129Luoyu Road,Wuhan 430079,China;²The First Institute of Oceanography,SOA,Qingdao 266061,China;³Guangdong Electric Power Research Institute,Guangzhou,China

Funds: The National Basic Research Program of China , N o. 2012CB725301; the National Natural Science Foundation of China, No. 41071268; Doctoral Fund of Ministry of Education of China, No. 20120141110035; Key Scientific and Technological Project of China Southern Power Grid, No. K-GD2013-030.

More Information

Author Bio:
XU Wenxue,PhD,specializes in LiDAR point clouds processing and 3Dmodeling.

Corresponding author:
GAO Wenwu

摘要

摘要: 目的提出了利用标记点过程从机载激光扫描数据中直接提取建筑物的方法。该方法首先根据建筑物在点云中的几何特征建立Gibbs能量模型,通过目标的一致性建立模型的数据项,通过目标的拓扑性质等空间特性建立模型的先验项;然后,利用可逆跳转马尔科夫蒙特卡洛算法(RJMCMC)和模拟退火算法优化求解;最后,利用精细处理移除错误提取的地面点、噪声点和树木点,合并相邻的目标,实现建筑物目标的精确提取。利用3组ISPRS机载激光扫描点云进行实验,结果表明,该方法能够准确、有效地提取建筑物,具有较强的稳健性。
- 建筑物 /
- 标记点过程 /
- Gibbs能量模型 /
- 可逆跳转马尔科夫蒙特卡洛 /
- 模拟退火 /
- 精细处理
Abstract: Objective In this paper,a marked point process based method is used to extract buildings from air-borne LIDAR data.At first,a Gibbs energy model is build according to the geometric feature of theobject in the point cloud data.This model contains both a data coherence term which fits the objects tothe data and a prior term which incorporates the prior knowledge of the object geometric properties.Then the previously defined model is optimized by the RJMCMC(Reverse Jump Markov Chain MonteCarlo)algorithm and simulated annealing algorithm.Finally,fine processing removes the terrestrialpoints,noise points and tree crown points of the extracted objects which are mistakenly extracted asbuildings,while combining adjacent objects.The method was tested with three different aerial LiDARdata sets from ISPRS.The results show that our method is capable of efficient and robust building ex-traction.
- building /
- marked point process /
- Gibbs energy model /
- RJMCMC /
- simulated annealing /
- fine processing