机载LiDAR点云中精细建筑物顶面的渐进提取  

Progressively Extracting Accurate Building Roofs from Airborne LiDAR Data

作　　者：赵瑞斌[1,2] 庞明勇[1] 张燕玲[2] 魏明强[3] 游国忠[4] Zhao Ruibin;Pang Mingyong;Zhang Yanling;Wei Mingqiang;You Guozhong(Department of Educational Technology, Nanjing Normal University, Nanjing 210097;Anhui Center for Collaborative Innovation in Geographical Information Integration and Application, Chuzhou University, Chuzhou 239000;College of Computer Science & Technology, Nanjing University of Aeronautics & Astronautics, Nanjing 210016;Department of Computer Science and Information Engineering, Aletheia University, Taipei 25103)

机构地区：[1]南京师范大学教育技术系,南京210097 [2]滁州学院安徽地理信息集成应用协同创新中心,滁州239000 [3]南京航空航天大学计算机科学与技术学院,南京210016 [4]真理大学咨询工程系,台北25103

出　　处：《计算机辅助设计与图形学学报》2017年第4期624-631,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：安徽省高校自然科学基金重点项目(KJ2014A184);安徽地理信息集成应用协同创新中心重点项目(201116Z01);江苏省社会科学基金项目(15TQB005);江苏省教育技术研究课题(2014R33356);滁州学院规划项目(2015GH21)

摘　　要：由LiDAR点云数据准确提取建筑物顶面是实现三维建筑模型自动重建的关键步骤.在分析现有顶面提取方法的基础上,提出一种渐进地提取LiDAR点云数据中精细建筑物顶面的方法.先以法向阈值和曲率阈值为约束,借助区域生长算法对原始点云进行初步分割,并得到面积较大、边界特征较明显的初始顶面;再借助主元分析法估算每个初始顶面的平面方程,并以点到平面的距离为约束,利用基于距离的区域生长算法提取其对应的精确顶面;最后通过随机抽样一致性算法(RANSAC)迭代地提取剩余点云中的小顶面.实验表明,通过动态调整阈值和迭代步骤,能够从LiDAR数据中精确地提取出复杂建筑物的顶面.Extracting building roofs from LiDAR data is a key processing step of3D building model reconstruction.In this paper,we present a progressive method for accurately extracting building roofs with complexshapes from airborne LiDAR data.Our method first segments original LiDAR data into a set of roughroofs,with larger area and distinct edges,by the region growing algorithm with a normal threshold and acurvature threshold.Accurate roofs are extracted based on estimating plane equations of the rough roofs usingthe principal component analysis technique.Our method finally employs the Random Sample Consensusalgorithm(RANSAC)to extract smaller roofs from the LiDAR data removed the points belong to the extractedroofs.Experimental results show that the method can robustly extract accurate building roofs in aprogressive way from the sampled data by adjusting the thresholds dynamically.

关 键 词：建筑物顶面 LIDAR 点云 渐进提取 区域生长 RANSAC 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]