测绘通报 ›› 2021, Vol. 0 ›› Issue (9): 15-20,27.doi: 10.13474/j.cnki.11-2246.2021.0266

• 学术研究 • 上一篇    下一篇

三维点云室内平面要素提取与优化

宋锦阳1, 钱建国1, 汤圣君2,3,4,5, 王伟玺2,3,4,5, 李晓明2,3,4,5   

  1. 1. 辽宁工程技术大学测绘与地理科学学院, 辽宁 阜新 123000;
    2. 深圳大学建筑与城市规划学院智慧城市研究院, 广东 深圳 518061;
    3. 自然资源部城市国土资源监测与仿真重点实验室, 广东 深圳 518061;
    4. 深圳市空间信息智能感知与服务重点实验室, 广东 深圳 518061;
    5. 广东省城市空间信息工程重点实验室, 广东 深圳 518061
  • 收稿日期:2020-09-01 出版日期:2021-09-25 发布日期:2021-10-11
  • 作者简介:宋锦阳(1995-),男,硕士生,研究方向为室内三维重构和激光点云数据处理。E-mail:543603701@qq.com
  • 基金资助:
    自然资源部城市国土资源监测与仿真重点实验室开放基金(KF-2019-04-010);国家自然科学基金(41801392;41901329;41971354;41971341);深圳市科技计划基础研究(自由探索)(JCYJ20180305125131482);深圳大学高水平大学建设2期(000002110335)

3D point cloud indoor plane element extraction and optimization methodology

SONG Jinyang1, QIAN Jianguo1, TANG Shengjun2,3,4,5, WANG Weixi2,3,4,5, LI Xiaoming2,3,4,5   

  1. 1. School of Geomatics, Liaoning Technical University, Fuxin 123000, China;
    2. Institute of Smart City, School of Architecture and Urban Planning, Shenzhen University, Shenzhen 518061, China;
    3. Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources, Shenzhen 518061, China;
    4. Key Laboratory of Spatial Information Intelligent Perception and Service, Shenzhen 518061, China;
    5. Key Laboratory of Urban Spatial Information Engineering, Guangdong Province, Shenzhen 518061, China
  • Received:2020-09-01 Online:2021-09-25 Published:2021-10-11

摘要: 室内平面要素的准确提取与关系恢复是室内模型自动化语义重建的重要基础,本文提出了一种面向复杂三维点云的室内平面要素提取与优化方法。该方法首先利用区域增长和RANSAC平面混合分割方法分割室内点云数据;其次利用室内语义部件的空间位置信息及包围盒和法向量信息,对分割后的平面进行平面要素的精确提取;然后对提取的墙面进行优化,实现共享墙面的合并,解决室内墙面冗余的问题;最后利用门与墙的空间位置信息,恢复门墙关联关系。试验部分采用了两组试验数据:一组是深圳大学某层教学楼的激光点云数据,另一组是国际摄影测量与遥感学会(ISPRS)的标准数据,通过对试验结果进行评估,验证了本文方法的有效性和可靠性。

关键词: 室内点云, 平面混合分割, 平面要素提取, 共享墙面合并, 门墙关联关系

Abstract: The accurate extraction and relationship restoration of indoor plane elements is an important basis for the automatic semantic reconstruction of indoor models. This paper proposes an indoor plane element extraction and optimization method for complex three-dimensional point clouds, which firstly uses area growth and RANSAC planar hybrid segmentation methods to segment indoor point cloud data; and then uses spatial location information of indoor semantic components as well as bounding box and normal vector information, the planar elements of the segmented plane are extracted accurately; then the extracted walls are optimized to achieve the merging of shared walls and solve the problem of indoor wall redundancy; finally, the spatial location information of doors and walls is used to restore the door-wall association relationship. In the experimental part, two sets of experimental data are used:one set is the laser point cloud data of a teaching floor of Shenzhen University, the other set is the standard data of International Society for Photogrammetry and Remote Sensing (ISPRS), and the validity and reliability of the method of this paper are verified by evaluating the experimental results.

Key words: indoor point clouds, plane hybrid segmentation, plane element extraction, shared wall merging, door-wall correlations

中图分类号: