测绘学报 ›› 2015, Vol. 44 ›› Issue (9): 980-987.doi: 10.11947/j.AGCS.2015.20140339

• 摄影测量学与遥感 • 上一篇    下一篇

车载激光扫描数据中多类目标的层次化提取方法

董震1,2, 杨必胜1,2   

  1. 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. 武汉大学时空数据智能获取技术与应用教育部工程研究中心, 湖北 武汉 430079
  • 收稿日期:2014-07-01 修回日期:2014-12-25 出版日期:2015-09-24 发布日期:2015-09-24
  • 作者简介:董震(1988—),男,博士生,研究方向为激光扫描数据处理。E-mail:dongzhenwhu@whu.edu.cn
  • 基金资助:
    国家973计划(2012CB725301);国家自然科学基金(41071268)

Hierarchical Extraction of Multiple Objects from Mobile Laser Scanning Data

DONG Zhen1,2, YANG Bisheng1,2   

  1. 1. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China;
    2. Engineering Research Center for Spatial-temporal Data Smart Acquisition and Application, Ministry of Education of China, Wuhan University, Wuhan 430079, China
  • Received:2014-07-01 Revised:2014-12-25 Online:2015-09-24 Published:2015-09-24
  • Contact: 杨必胜,bshyang@whu.edu.cn E-mail:bshyang@whu.edu.cn
  • Supported by:
    The National Basic Research Program of China(973 Program)(No.2012CB725301);The National Natural Science Foundation of China (No.41071268)

摘要: 提出了一种从车载激光扫描数据中层次化提取多类型目标的有效方法。该方法首先利用颜色、激光反射强度、空间距离等特征,生成多尺度超级体素;然后综合超级体素的颜色、激光反射强度、法向量、主方向等特征利用图分割方法对体素进行分割;同时计算分割区域的显著性,以当前显著性最大的区域为种子区域进行邻域聚类得到目标;最后结合聚类区域的几何特性判断目标可能所属的类别,并按照目标类别采用不同的聚类准则重新聚类得到最终目标。试验结果表明,该方法成功地提取出建筑物、地面、路灯、树木、电线杆、交通标志牌、汽车、围墙等多类目标,目标提取的总体精度为92.3%。

关键词: 车载激光点云, 多尺度超级体素, 多类型目标提取, 显著性, 层次化提取

Abstract: This paper proposes an efficient method to extract multiple objects from mobile laser scanning data. The proposed method firstly generates multi-scale supervoxels from 3D point clouds using colors, intensities and spatial distances. Then, a graph-based segmentation method is applied to segment the supervoxels by integrating their colors, intensities, normal vectors, and principal directions. Then, the saliency of each segment is calculated and the most salient segment is selected as a seed to cluster for objects clustering. Hence, the objects are classified and the constraint conditions of object's category are included to re-clustering for more accurate extraction of objects. Experiments show that the proposed method has a promising solution for extracting buildings, ground, street lamps, trees, telegraph poles, traffic signs, cars, enclosures and the objects extraction overall accuracy is 92.3%.

Key words: mobile laser scanning, multi-scale supervoxel, multiple object extraction, saliency, hierarchical extraction

中图分类号: