支持向量机的建筑物激光脚点提取方法Support vector machine classification for detecting laser points of building
罗伊萍,姜挺,王鑫,陈文锋,张锐
摘要(Abstract)：
本文提出了一种基于全色波段航空影像和激光雷达数据的建筑物检测方法。如何从激光点云数据中提取出建筑物激光脚点,是建筑物三维重建和轮廓提取的难点问题之一。植被密集区域以及与建筑物紧密相邻的树木的激光点很难与建筑物激光点区分开。本文利用支持向量机对单个激光点的特征进行两分类,特征向量包括激光点的高程、高程变化信息以及与激光点配准的影像光谱信息。实验表明,基于支持向量机的点态分类算法能够有效提取建筑物激光脚点,影像光谱信息能明显提高分类精度。
关键词(KeyWords)： 激光雷达;支持向量机;建筑物检测;点态分类;影像特征
基金项目(Foundation):
作者(Author): 罗伊萍,姜挺,王鑫,陈文锋,张锐
DOI: 10.16251/j.cnki.1009-2307.2011.04.004
参考文献(References)：
- [1]FILIN S.Surface clustering from airborne laser scanningdata[J].IAPRS.Graz,Austria,2002,34/3 A:119-124.
- [2]ROTTENSTEINER F,TRINDER J,CLODE S,etal.Using the Dempster Shafer method for the fusion ofLiDar data and multi-spectral images for building detec-tion[J].Inf.Fusion.2004,6(4).
- [3]MORGAN H,HABIB A.Interpolation of LIDAR data andautomatic building extraction[C]//ACSM-ASPRS AN-NUAL CONFERENCE PROCEEDINGS.2002:85-91.
- [4]ZHANG K,YAN J,CHEN S C.Automatic Construc-tion of Building Footprints from Airborne LiDAR Data[C]//IEEE Transaction on Geoscience and RemoteSensing.2006,44(9).
- [5]ELBERINK S O,MASS H G.The use of anisotropicheight texture measures for the segmentation of airbornelaser scanner data[C]//International Archives of Pho-togrammetry and Remote Sensing.Amsterdam,Nether-lands.2000,XXXIII,Part B3:678-684.
- [6]ALHARTHY A,BETHEL J.Heuristic filtering and 3Dfeature extraction from LIDAR data[J].IAPRS.Graz,Austria,2002,33:29-35.
- [7]MA R.DEM Generation and Building Detection from Li-dar Data[J].Photogrammetry Engineering&RemoteSensing,2005,71(3).
- [8]SOHN G,DOWMAN I.Data fusion of high-resolutionsatellite imagery and LiDAR data for automatic buildingextraction[J].ISPRS Journal of Photogrammetry&Remote Sensing.2007,62(2).
- [9]LEENA M,HARRI K,JUHA H.Classification Treebased Building Detection From Laser Scanner and AerialImage Data[C]//IAPRS.Finland,2007,September12-14:Volume XXXVI,Part 3/W52:280-287.
- [10]SECORD J,ZAKHOR A.Tree Detection in Urban Re-gions Using Aerial Lidar and Image Data[C]//IEEEGEOSCIENCE AND REMOTE SENSING LETTERS,2007,4(2):196-200.
- [11]张峰,等.基于SVM的多源遥感影像面向对象建筑物提取方法[J].国土资源遥感,2008,(2).
- [12]周绮凤.基于支持向量机的若干分类问题研究[D].厦门:厦门大学,2007.
- [13]ZHANG K,CHEN S C,WHITMAN D,et al.A Pro-gressive Morphological Filter for Removing NongroundMeasurement from LIDAR Data[C]//IEEE Transac-tions on Geoscience and Remote Sensing.2003,41(4).
- [14]张皓.机载LIDAR数据滤波及建筑物提取技术研究[D].郑州:信息工程大学,2009.