Abstract：The present paper proposes a novel hyperspectral image classification algorithm based on LS-SVM (least squares support vector machine). The LS-SVM uses the features extracted from subspace of bands (SOB). The maximum noise fraction (MNF) method is adopted as the feature extraction method. The spectral correlations of the hyperspectral image are used in order to divide the feature space into several SOBs. Then the MNF is used to extract characteristic features of the SOBs. The extracted features are combined into the feature vector for classification. So the strong bands correlation is avoided and the spectral redundancies are reduced. The LS-SVM classifier is adopted, which replaces inequality constraints in SVM by equality constraints. So the computation consumption is reduced and the learning performance is improved. The proposed method optimizes spectral information by feature extraction and reduces the spectral noise. The classifier performance is improved. Experimental results show the superiorities of the proposed algorithm.
Key words：Subspace of bands;Maximum noise fraction;Least squares;Feature extraction
[1] Guo B, Gunn S R, Damper R I, et al. IEEE Transactions On Image Processing, 2008, 17(4): 622. [2] Chen J, Wang C, Wang R. IEEE Trans. Geosci. Remote Sens., 2009, 47(7): 2193. [3] Chen J, Wang C, Wang R. Neurocomputing, 2009, 72: 3370. [4] Vapnik V N. Statistical Learning Theory. New York: Wiley, 1998. [5] Suykens J A K, Vandewalle L. Neural Processing Letters, 1999, 9(3): 293. [6] Jia X,!Richards J A. IEEE Trans. Geosci. Remote Sens, 1999, 37(1): 538. [7] Green A, Berman M, Switzer P, et al. IEEE Trans. Geosci. Remote Sens., 1988, 26(1): 65. [8] Xin Q, Nian Y, Li X, et al. Journal of Electronics (China), 2009, 26(6): 831. [9] Melgani F. Bruzzone L. IEEE Trans. Geosci. Remote Sens., 2004, 42(8): 1778. [10] David C H. Journal of Machine Learning Research, 2008, 9: 2733. [11] Plaza A, Benediktsson J A, Boardman J W, et al. Remote Sensing of Environment, 2009, 113: 110. [12] Mattia M, Camps-Valls G, Bruzzone L. IEEE Geoscience and Remote Sensing Letters, 2009, 6(2): 234. [13] Barat M, Hamid A-M, Mohammad J V Z, et al. IEEE Trans. Geosci. Remote Sens., 2009, 47(7): 2091. [14] ftp://ftp.ecn.purdue.edu/biehl/MultiSpec/92AV3C/. [15] Tarabalka Y, Chanussot J, Benediktsson J A. Pattern Recognition, 2010, 43: 2367.
