A NOVEL OVER-SEGMENTATION METHOD FOR POLARIMETRIC SAR IMAGES CLASSIFICATION

被引:11
作者
He, Chu [1 ]
Deng, Jingbo
Xu, Lianyu [1 ]
Li, Shuang [1 ]
Duan, Mengmeng
Liao, Mingsheng
机构
[1] Wuhan Univ, Sch Elect Informat, Wuhan, Peoples R China
来源
2012 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS) | 2012年
关键词
D O I
10.1109/IGARSS.2012.6351717
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
This paper, we propose a novel over-segmentation method Feature Geometry Space Fusion (FGSF) for polarimetric SAR (POLSAR) data classification, which uses the polarimetric feature and geometric feature. In order to exam its performance, experiments on the data acquired by AIRSAR show that the over segment regions segmented by FGSF method performs better than mean shift when used for classification.
引用
收藏
页码:4299 / 4302
页数:4
相关论文
共 9 条
[1]  
Achanta R., 2010, TECHNICAL REPORT, P2
[2]   Using functional domain composition and support vector machines for prediction of protein subcellular location [J].
Chou, KC ;
Cai, YD .
JOURNAL OF BIOLOGICAL CHEMISTRY, 2002, 277 (48) :45765-45769
[3]   Power watersheds: A new image segmentation framework extending graph cuts, random walker and optimal spanning forest [J].
Couprie, Camille ;
Grady, Leo ;
Najman, Laurent ;
Talbot, Hugues .
2009 IEEE 12TH INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2009, :731-738
[4]  
Feng Lin, 2008, Chinese Journal of Scientific Instrument, V29, P649
[5]  
Han Zhi, 2005, THESIS
[6]  
Kim J, 2011, PROC CVPR IEEE, P1553, DOI 10.1109/CVPR.2011.5995526
[7]  
Lee JS, 2009, OPT SCI ENG-CRC, P1
[8]  
Li Kun, 2010, Proceedings of the 2010 International Conference on Electrical and Control Engineering (ICECE 2010), P1399, DOI 10.1109/iCECE.2010.346
[9]   Normalized cuts and image segmentation [J].
Shi, JB ;
Malik, J .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2000, 22 (08) :888-905