For Polarimetric SAR (PolSAR), because it contains more scattering information, thus it can provide more available features. How to use the features is crucial for the PolSAR image classification, however, there are no existing specific rules. To solve the above problem, a supervised Polarimetric SAR image classification method via weighted ensemble based on 0-1 matrix decomposition is proposed. The proposed method adopts matrix decomposition ensemble to learn on different feature subsets to get coefficients, and weighting ensemble algorithm is employed via the predictive results to improve the final classification results. Firstly, some features are extracted from PolSAR data as initial feature group and are divided randomly into several feature subsets. Then, according to the ensemble algorithm to get the different weights based on the feature subsets, small coefficients are assigned to bad classification results to decrease the harmful impact of some features. The final classification result is achieved by combining the results together. The experimental results of L-band and C-band PolSAR data demonstrate that the proposed method can effectively improve the classification results.
Lee J S and Pottier E. Polarimetric Radar Imaging from Basic to Application[M]. New York: CRC Press, 2011: 1-51, 66-72, 229-247.
[2]
Ding Tao, Anfinsen S N, and Brekke C. A comparative study of sea clutter covariance matrix estimators[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(5): 1010-1014.
[3]
Cloude S R and Pottier E. A review of target decomposition theorems in radar polarimetry[J]. IEEE Transactions on Geoscience and Remote Sensing, 1996, 34(2): 498-518.
[4]
Wang Chun-le, Yu Wei-dong, Wang Robert, et al.. Comparison of nonnegative eigenvalue decompositions with and without reflection symmetry assumptions[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(4): 2278-2286.
[5]
Zhang Hong, Xie Lei, Wang Chao, et al.. Investigation of the capability of H-decomposition of compact Polarimetric SAR[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(4): 868-872.
[6]
Freeman A and Durden S L. A three-component scattering model for polarimetric SAR[J]. IEEE Transactions on Geoscience and Remote Sensing, 1998, 36(3): 963-973.
[7]
Jiao Zhi-hao, Yang Jian, Yeh Chun-mao, et al.. Modified three-component decomposition methord for polarimetric SAR data[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(1): 200-204.
Liu Gao-feng, Li Ming, Wang Ya-jun, et al.. An improved adaptive non-negative eigenvalue decomposition for polarimetric systhetic aperture radar[J]. Journal of Electronics & Information Technology, 2013, 35(6): 1449-1455.
[9]
Yamaguchi Y, Moriyama T, Ishido M, et al.. Four-component scattering model for polarimetric SAR image decomposition [J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(8): 1699-1706.
[10]
An W, Xie C, Yuan X, et al.. Four-component decomposition of polarimetric SAR image with deorientation[J]. IEEE Geoscience and Remote Sensing Letters, 2011, 8(6): 1090-1094.
Liu Gao-feng, Li Ming, Wang Ya-jun, et al.. Yamaguchi decomposition based on hierarchical nonnegative eigenvalue restriction[J]. Journal of Electronics & Information Technology, 2013, 35(11): 2678-2685.
[12]
Liu Gao-feng, Li Ming, Wang Ya-jun, et al.. Four-component scattering power decomposition of remainder coherency matrices constrained for nonnegative eigenvalues[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(2): 494-498.
[13]
Lee J S, Grunes M R, and Kwok T. Classification of multi-look polarimetric SAR imagery based on complex Wishart distribution[J]. International Journal of Remote Sensing, 1994, 15(11): 2299-2311.
[14]
Fukuda S and Hirosawa H. Support vector machine classification of land cover: application to polarimetric SAR data[C]. IEEE International Geoscience And Remote Sensing Symposium (IGARSS’01), Sydney, Australia, 2001: 187-189.
[15]
Fukuda S, Katagiri R, and Hirsosawa H. Unsupervised approach for polarimetric SAR image classification using support vector machines[C]. IEEE International Geoscience And Remote Sensing Symposium (IGARSS’02), Toronto, Canada, 2002, 5: 2599-2601.
[16]
Kumar S, Ghosh J, and Crawford M. Best-bases feature extraction algorithm for classification of hyperspectral data[J]. IEEE Transactions on Geoscience and Remote Sensing, 2001, 39(7): 1368-1379.
Xu Feng and Jin Ya-qiu. Theory and application of deorientation for target scattering Part I: deorientation theory[J]. Chinese Journal of Radio Science, 2006, 21(1): 6-15.
[18]
Mao Sha-sha, Xiong Lin, Jiao Li-cheng, et al.. Weighted ensemble based on 0-1 matrix decomposition[J]. Electronics Letters, 2013, 49(2): 116-118.
[19]
Henri Maitre. 孙洪, 等译. 合成孔径雷达图像处理[M]. 北京: 电子工业出版社, 2005: 第4章.
[20]
Kuncheva L I. Combining Pattern Classifiers: Methods and Algorithms[M]. Hoboken, New Jersey, John Wiley & Sons, 2004: Chapter 1.
[21]
Christopher J C B. A tutorial on support vector machine for pattern recognition[J]. Data Mining and Knowledge Discovery, 1998, 2(2): 121-167.
