Please wait a minute...
img

官方微信

遥感技术与应用  2014, Vol. 29 Issue (2): 330-337    DOI: 10.11873j.issn.1004-0323.2014.2.0330
图像与数据处理     
基于中间层特征的全极化SAR监督地物分类
任俊英,苏彩霞,曹永锋
(贵州师范大学数学与计算机科学学院,贵州 贵阳 550000)
Supervised Land\|use Classification of Full Polarization SAR Image based on Middle Level Feature
Ren Junying,Su,Caixia,Cao Yongfeng
(School of Mathematics and Computer Science,Guizhou Normal University,Guiyang 550000,China)
 全文: PDF(6131 KB)  
摘要:

提出了一种组合中间层特征(Middle Level Feature,MLF)和支持向量机(Support Vector Machine,〖JP2〗SVM)的全极化合成孔径雷达(Synthetic Aperture Radar,SAR)监督地物分类方法。选择监督方法的目的是直接区分实际地物类别,中间层特征在非监督聚类结果中获取,用于跨越底层特征与地物类别间的语义鸿沟。统计以某像素为中心的特征支持区域内各“中间成分”的占比作为该像素的MLF。这里“中间成分”对应于基于底层极化特征得到的非监督聚类类别。在覆盖武汉地区的Radarsat\|2全极化数据上,与基于经典全极化特征的SVM监督分类方法进行了对比,研究了不同中间成分获取方法以及特征支持窗口对于分类性能的影响,结果显示:该方法有很好的性能并有进一步提升的空间。〖JP〗

关键词: 全极化SAR支持向量机中间层特征中间成分    
Abstract:

A supervised method combining Middle Level Feature (MLF) with Support Vector Machine (SVM) was proposed for land\|use classification of full polarization Synthetic Aperture Radar (SAR) image.Supervised method is chosen to directly distinguish the actual land-use categories.The MLF that is used for striding over the semantic gap between the low-level polarization scattering characteristics and the high\|level semantics of land-use categories is got from the result of classic unsupervised classification methods for full polarization SAR image.The MLF of a pixel is calculated by counting the frequency of the middle-components in a feature supporting region centered on the pixel.Here the middle-components refer to the unsupervised clustering categories obtained from the low\|level polarization characteristics.The proposed method is tested on a Radarsat-2 full polarization data covering WUHAN area and good classification performance and potential of further improvement are shown.The comparison between the supervised classification method combining SVM and the classic polarization characteristics is given.The proposed method and different methods for getting the middle-components and feature supporting windows with different size are studied on their impact on the final classification performance.

Key words: Full polarization SAR    Support Vector Machine    Middle Level Feature    Middle-components
收稿日期: 2013-03-19 出版日期: 2014-05-14
:  TP 722.6  
基金资助:

国家自然科学基金“全极化SAR异质场景散射基元统计谱建模与分类”(41161065),“高分辨率SAR图像复杂场景建模与基于场景的目标检测”(40901207),贵州省科学技术厅,贵州师范大学联合科技基金资助项目“基于SAR信息技术的贵州水稻估产研究”(黔科合J字LKS[2013]28号),贵州师范大学学生科研基金。

通讯作者: 曹永锋(1976-),男,河北冀州人,教授,主要从事SAR图像处理与解译研究。Email:yongfengcao.cyf@gmail.com。    
作者简介: 任俊英(1987-),女,河南濮阳人,硕士研究生,主要从事遥感图像计算机处理与解译方面的研究。Email:renjunying163@163.com。
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  

引用本文:

任俊英,苏彩霞,曹永锋. 基于中间层特征的全极化SAR监督地物分类[J]. 遥感技术与应用, 2014, 29(2): 330-337.

Ren Junying,Su,Caixia,Cao Yongfeng. Supervised Land\|use Classification of Full Polarization SAR Image based on Middle Level Feature. Remote Sensing Technology and Application, 2014, 29(2): 330-337.

链接本文:

http://www.rsta.ac.cn/CN/10.11873j.issn.1004-0323.2014.2.0330        http://www.rsta.ac.cn/CN/Y2014/V29/I2/330

[1]Wang Chao,Zhang Hong,Chen Xi,et al.Full Polarization Synthetic Aperture Radar[M].Bejing:Science Press,2008.[王超,张红,陈曦,等.全极化合成孔径雷达[M].北京:科学出版社,2008.] 

[2]Kong J A,Yueh S H,Shin R T,et al.Classification of Earth Terrain Using Polarimetric Synthetic Aperture Radar Images[C]//Kong J A.PIER.Amsterdam,the Netherlands:Elsevier,1990.

[3]Cloude S R,Pottier E.An Entropy based Classification Scheme for Land Applications of Polarimetric SAR[J].IEEE Transactions on Geoscience and Remote Sensing,1997,35(1):68-78.

[4]Freeman A,Durden S.A Three-component Scattering Model for Polarimetric SAR Data[J].IEEE Transactions on Geoscience and Remote Sensing,1998,36(3):963-973.

[5]Cao F,Hong W.A New Classification Method based on Cloude-pottier Eigenvalue/eigenvector Decomposition[C]//IEEE International Geoscience and Remote Sensing Symposium,2005,1:296-299.

[6]Wang S,Liu K,Pei J J,et al .Unsupervised Classification of Fully Polarimetric SAR Images based on Scattering Power Entropy and Copolarized Ratio[J].IEEE Geoscience and Remote Sensing Letters,2013,10(3):622-626.

[7]Lee J S,Grunes M R,Ainsworth T L,et al .Classification of Multi-look Polarimetric SAR Imagery based on Complex Wishart Distribution[J].IEEE Transactions on Geoscience and Remote Sensing,1999,37(5):2249-2258.

[8]Pottier E.Unsupervised Classification Scheme and Topography Derivation of PolSAR Data based on the H/A/α Polarmetric Decomposition Theorem[C]//Proceedings of the 4th International Workshop on Radar Polarimetry,Nantes,France,1998.

[9]Lee J S,Mitchell R G,Pottier E,et al .Unsupervised Terrain Classification Preserving Polarimetric Scattering Characteristics[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(4):722-731.

[10]Lee J S,Mitchell R G,kwok r.Classification of Multi-look Polarization SAR Imagery based on the Complex Wishart Distribution[J].International Journal of Remote Sensing,1994,15(11):2299-2311.

[11]Chen Jinsong,Shao Yun,Li Zhen.Full Polarization SAR Neural Network Classification based on the Theory of Target Decomposition[J].Chinese Journal of Image and Graphics,2004,9(5):552-556.[陈劲松,邵云,李震.基于目标分解理论的全极化SAR 图像神经网络分类方法[J].中国图象图形学报,2004,9(5):552-556.]

[12]Tian Xin,Chen Erxue,Li Zengyuan,et al.Rice/Dry-land Crop Discrimination Using Multi-polarization Satellite SAR Data——A Case Study in Haian County of Jiangsu Province[J].Remote Sensing Technology and Application,2012,27(3):406-412.[田昕,陈尔学,李增元,等.基于多极化星载SAR数据的水稻/旱田识别——以江苏省海安县为例[J].遥感技术与应用,2012,27(3):406-412.]

[13]Corr D G,Cloude S R,Ferro-Famil L,et al.A Review of the Application of  SAR Polarimetry and  Polarimetric Interfe- rometry——An ESA Funded Study[C]//Proceedings of the Workshop on POLinSAR-Applications of SAR Polarimetry and Polarimetric Interferometry (ESA SP-529),2003,Frascati,Italy.

[14]Yang W,Liu Y,Xia G S,et al.Statistical Mid-level Features for Building-up Area Extraction from Full Polarimetric SAR Imagery[J].Progress in Electromagnetics Research,2012,132:233-254.

[15]GB 50137-2011.Standard of Urban Land Classification and Planning Construction Land[S].2011.[ GB 50137-2011.城市用地分类与规划建设用地标准[S].2011.]

[16]Andrew R.Webb,Statistical Pattern Recognition,Second Edition[M].Beijing:Publishing House of Electronics Industry,2004.


 

[1] 李姣姣,刘玉,陈锟山. 基于香农熵的极化SAR相干矩阵信息量评价#br#[J]. 遥感技术与应用, 2018, 33(5): 842-849.
[2] 熊伟,徐永力,姚力波,崔亚奇. 基于SVM的高分辨率SAR图像舰船目标检测算法[J]. 遥感技术与应用, 2018, 33(1): 119-127.
[3] 陈洋波,张涛,窦鹏,董礼明,陈华. 基于SVM的东莞市土地利用/覆被自动分类误差来源与后处理[J]. 遥感技术与应用, 2017, 32(5): 893-903.
[4] 史飞飞,高小红,杨灵玉,何林华,贾伟. 基于HJ-1A高光谱遥感数据的湟水流域典型农作物分类研究[J]. 遥感技术与应用, 2017, 32(2): 206-217.
[5] 胡根生,吴问天,黄文江,梁栋,黄林生. 粒子群优化的最小二乘支持向量机在小麦白粉病监测中的应用[J]. 遥感技术与应用, 2017, 32(2): 299-304.
[6] 翟玮,沈焕锋,黄春林. 结合PolSAR影像纹理特征分析提取倒塌建筑物[J]. 遥感技术与应用, 2016, 31(5): 975-982.
[7] 鲍蕊,夏俊士,薛朝辉,杜培军,车美琴. 基于形态学属性剖面的高光谱影像集成分类[J]. 遥感技术与应用, 2016, 31(4): 731-738.
[8] 桂容,徐新,董浩,宋超. 全极化SAR影像城区建筑密度分析[J]. 遥感技术与应用, 2016, 31(2): 267-274.
[9] 林志垒,晏路明. 高光谱影像的BDT-SVM地物分类算法与应用[J]. 遥感技术与应用, 2016, 31(1): 177-185.
[10] 尹小君,张清,赵庆展,汪传建,宁川. 基于SVM的加工番茄细菌性斑点病氮素含量反演[J]. 遥感技术与应用, 2015, 30(3): 461-468.
[11] 周艳飞,张绘芳,李霞,杨帆,丁程锋. 基于高分辨遥感数据的胡杨与柽柳树冠提取[J]. 遥感技术与应用, 2015, 30(3): 510-517.
[12] 邹利东,郭航,朱秀芳,王艳艳,郑学昌,王建阔. 设施农业空间分布信息自动提取方法研究[J]. 遥感技术与应用, 2014, 29(4): 669-674.
[13] 周颂洋,谭琨,吴立新. 基于邻域距离ISOMAP算法的高光谱遥感降维算法[J]. 遥感技术与应用, 2014, 29(4): 695-700.
[14] 刘朝相,宫兆宁,赵文吉. 基于SVM模型的妫水河叶绿素a浓度的遥感反演[J]. 遥感技术与应用, 2014, 29(3): 419-427.
[15] 张汉奎,黄 波,俞 乐. SVM-RFE高光谱数据波段选择中核函数的研究[J]. 遥感技术与应用, 2013, 28(5): 747-752.