一种结合分水岭与决策树C5.0的极化SAR分类方法

张剑清, 段艳

张剑清, 段艳. 一种结合分水岭与决策树C5.0的极化SAR分类方法[J]. 武汉大学学报 ( 信息科学版), 2014, 39(8): 891-896. DOI: 10.13203/j.whugis20120112
引用本文: 张剑清, 段艳. 一种结合分水岭与决策树C5.0的极化SAR分类方法[J]. 武汉大学学报 ( 信息科学版), 2014, 39(8): 891-896. DOI: 10.13203/j.whugis20120112
ZHANG Jianqing, DUAN Yan. A Supervised Classification Method of Polarimetric Sythetic ApertureRadar Data Using Watershed Segmentation and Decision Tree C5.0[J]. Geomatics and Information Science of Wuhan University, 2014, 39(8): 891-896. DOI: 10.13203/j.whugis20120112
Citation: ZHANG Jianqing, DUAN Yan. A Supervised Classification Method of Polarimetric Sythetic ApertureRadar Data Using Watershed Segmentation and Decision Tree C5.0[J]. Geomatics and Information Science of Wuhan University, 2014, 39(8): 891-896. DOI: 10.13203/j.whugis20120112

一种结合分水岭与决策树C5.0的极化SAR分类方法

基金项目: 国家973计划资助项目(2012CB719904)
详细信息
    作者简介:

    张剑清,教授,现主要从事摄影测量与遥感、计算机视觉研究。

  • 中图分类号: P237.3

A Supervised Classification Method of Polarimetric Sythetic ApertureRadar Data Using Watershed Segmentation and Decision Tree C5.0

Funds: The National Program on Key Basic Research Project of China(973Program),No.2012CB719904.
More Information
    Author Bio:

    ZHANG Jianqing,professor. His research fields include photogrammetry,remote sensing and computer vision.

  • 摘要: 目的 提出了一种利用多种极化特征并结合分水岭算法与决策树C5.0分类器的极化SAR数据分类方法。首先对极化SAR数据进行极化精致Lee滤波,接着对其进行极化分解得到多个极化通道与 Pauli RGB图像,改进梯度图生成法并进行形态学分水岭分割与区域合并,最后选择样本构建决策树 C5.0分类器并进行分类。实验结果表明,该方法与传统基于像素的分类方法相比精度有显著提高,同时由于使用了较多的极化特征,也使分类精度在一定程度上得到了提高。
    Abstract: Objective A supervised classification method of polarimetric sythetic aperture radar(PoSAR)data u-sing watershed segmentation and Decision Tree C5.0with many polarimetric channels is proposed.First,the PolSAR data was filtered using the 5×5refined Lee PolSAR speckle filter,and then a PauliRGB color image and many polarimetric channels were obtained using various algorithms.Then,wa-tershed segmentation on gradient map was made for a homogeneous area and the features of every areawere worked out.At last,Decision tree C5.0was used to deal with the data.The result shows thatthis method performs better than methods based on pixels,and the classification accuracy is improvedwith the quantity of polarimetric characteristic increase.
  • [1] Kouskoulas Y,Ulaby F T,Pierce L E.The Bayes-ian Hierarchical Classifier(BHC)and Its Applica-tion to Short Vegetation Using Multifrequency Po-larimetric SAR[J].IEEE Transactions on Geosci-ence and Remote Sensing,2004,42(2):469-477[2] Cloude S R,Pottier E.An Entropy Based Classifica-tion Scheme for Land Applications of PolarimetricSAR[J].IEEE Transactions on Geoscience and Re-mote Sensing,1997,35(1):68-78[3] Lang Fengkai,Yang Jie,Zhao Lingli,et al.Polari-metric SAR Data Classification with Freeman Entro-py and Anisotropy Analysis[J].Acta Geodaetica etCartographica Sinica,2012,41(4):556-562(郎丰铠,杨杰,赵伶俐,等.基于Freeman散射熵和各向异性度的极化SAR影像分类算法研究[J].测绘学报,2012,41(4):556-562)[4] Zou Tongyuan,Yang Wen,Dai Dengxin,et al.AnUnsupervised Classification Method of POLSARImage[J].Geomatic and Information Science ofWuhan University,2009,34(8):910-914(邹同元,杨文,代登信,等.一种新的极化SAR图像非监督分类算法研究[J].武汉大学学报·信息科学版,2009,34(8):910-914)[5] Wu Zhaocong,OuYang Qundong,Hu Zhongwen.Polarimetric SAR Image Classification Using Water-shed-transformation and Support Vector Machine[J].Geomatic and Information Science of Wuhan895武 汉 大 学 学 报 · 信 息 科 学 版2014年8月University,2012,37(1):7-12(巫兆 聪,欧 阳 群东,胡忠文.应用分水岭变换与支持向量机的极化SAR图像分类[J].武汉大学学报·信息科学版,2012,37(1):7-12)[6] Qi Z X,Yeh A G,Li X,et al.A Novel Algorithmfor Land Use and Land Cover Classification UsingRADARSAT-2Polarimetric SAR Data[J].RemoteSensing of Environment,2012,118:21-39[7] Lee J S,Grunes M R,Grandi G D.PolarimetricSAR Speckle Filtering and Its Implication for Clas-sification[J].IEEE Transactions on Geoscience andRemote Sensing,1999,37(5):2363-2373[8] Lee J S,Pottier E.Polarimetric Radar Imagingfrom Basics to Applications[M].New York:CRCPress,2009[9] Gonzalez R C,Woods R E.Digital Image Process-ing[M].2nd Edition.US:Prentice Hall,2002[10]Ye Qixiang,Gao Wen,Wang Weiqiang,et al.AColor Image Segmentation Algorithm by Using Col-or and Spatial Information[J].Journal of Soft-ware,2004,15(4):522-530(叶齐祥,高文,王伟强,等.一种融合颜色和空间信息的彩色图像分割算法[J].软件学报,2004,15(4):522-530)[11]Lawrence R L,Wright A.Rule-based ClassificationSystems Using Classification and Regression Tree(CART)Analysis[J].Photogrammetric Engineer-ing and Remote Sensing,2001,67(10):1137-1142
计量
  • 文章访问数:  1226
  • HTML全文浏览量:  58
  • PDF下载量:  584
  • 被引次数: 0
出版历程
  • 收稿日期:  2013-03-24
  • 修回日期:  2014-08-04
  • 发布日期:  2014-08-04

目录

    /

    返回文章
    返回