Feature Selection and Classification of Polarimetric SAR Images Using SVM

[1] Fukuda S and Hirosawa H. Support vector machineclassification of land cover: application to polarimetric SARdata. IEEE International Geoscience and Remote SensingSymposium, Sydney, Australia, July 2001: 187-189. [2] Cloude S R and Pottier E. An entropy based classificationscheme for land applications of polarimetric SAR[J].IEEETrans. on Geoscience and Remote Sensing.1997, 35(1):68-78 [3] Vapnik V N. 许建华, 张学工译. 统计学习理论. 北京: 电子工业出版社, 2004: 364-384. [4] Fukuda S, Katagiri R, and Hirosawa H. Unsupervisedapproach for polarimetric SAR image classification usingsupport vector machines. IEEE International Geoscience andRemote Sensing Symposium, Toronto, Canada, June 2002:2599-2601. [5] Liu Huan and Yu Lei. Toward integrating feature selectionalgorithms for classification and clustering[J].IEEE Trans. onKnowledge and Data Engineering.2005, 17(4):491-502 [6] Pudil P, Novovicova J, and Kittler J. Floating searchmethods in feature selection[J].Pattern Recognition Letters.1994, 15(11):1119-1125 [7] Narendra P M and Fukunaga K. A branch and boundalgorithm for feature subset selection[J].IEEE Trans. onComputers.1977, 26(9):917-922 [8] Kononenko I. Estimation attributes: analysis and extensionsof RELIEF. Proc. 7th European Conference on MachineLearning, Sicily, Italy, April 1994: 171-182.