基于子孔径的极化SAR图像目标分类算法研究
|
摘要
极化SAR能够同时获得目标区域的多通道与高分辨率数据,可更好的揭示目标的物理散射特性,因此,极化SAR图像的特征提取与目标分类在解释雷达图像和目标识别方面起着十分重要的作用。充分利用极化信息,可更加准确地理解目标散射机理,从而带来更好的SAR图像分类结果。本文根据极化SAR数据的特点,结合子孔径分析技术,进行了极化SAR图像目标分类算法的研究,主要工作如下:
1.总结了目前常用的极化SAR图像分类算法及其存在的问题。在此基础上,将AdaBoost算法与极化通道所携带的信息相结合,对极化SAR图像进行监督分类,在已知场景类别数的情况下,该方法能够加快收敛速度,改善分类效果;
2.研究了基于子孔径分析的极化散射机理与特征提取方法。首先从时频分解的角度对方位向子孔径进行了分析,对场景中存在的各向异性散射行为和布拉格谐振现象进行了讨论。针对场景中的非平稳目标,分析了已有的极大似然比非平稳目标检测算法,还研究了一种基于熵值与平均散射机理角的联合向量检测非平稳目标的方法。通过该方法可以对方位向频谱中的非平稳散射行为进行定位,最终消除非平稳散射在全孔径中的影响;
3.对目标的极化分解方法与基于极化分解的极化SAR图像分类方法进行了研究与仿真试验。如Pauli分解、Krogager分解、H/α分类、H/A/α分类、H/α/Wishart分类、H/A/α/Wishart分类和基于Freeman-Wishart的极化SAR图像分类等。在此基础上,改进了基于AdaBoost算法的极化SAR图像分类方法,将Pauli分解与AdaBoost算法的优势得以发挥,该方法既解决了AdaBoost算法需要知道场景先验知识的缺点,同时还能改善分类效果、提高收敛速度;
4.改进了基于全孔径数据的极化SAR图像分类方法,得到三种基于子孔径的极化SAR图像分类算法并进行了仿真试验:一是非平稳目标检测与H/α分类相结合的极化SAR图像分类;二是子孔径分解与H/α/Wishart迭代分类相结合的极化SAR图像分类;三是结合Freeman分解与子孔径散射特性的极化SAR图像分类。
仿真试验表明,结合子孔径分析与极化分解的极化SAR图像分类,能够在一定程度上改善分类效果、增加分类精细度、提高收敛速度,在实际的SAR图像分类应用中具有很大的价值。
Polarimetric SAR(PolSAR) combines the high space resolution of SAR system and multiple channels data of the targets, which can reveal the physical scattering characteristics better. Therefore, extraction of characteristics and objects classification of PolSAR image play an important part in interpretation of radar data and target recognition. Using polarimetric information fully can express scattering mechanism much more exactly, which can obtain much better result of SAR image. In this dissertation, based on characters of PolSAR data and subaperture analysis, we research on target classification of PolSAR image. The main work and contributions accomplished in the dissertation are as follows:
1. Firstly, the existing popular algorithms of PolSAR image classification and the problems are summarized. Besides, a new classification method, which employs AdaBoost algorithm, is proposed. Having known the types of objects in the scene, the classification accuracy is enhanced by using this method.
2. The subaperture analysis of PolSAR image is studied. Firstly, based on time-frequency analysis, the decomposition theory of subaperture is investigated. Anisotropy polarimetric behavior and Bragg-resonance are also researched, and the latter is one of the important reasons for azimuthal polarimetric changes. Secondly, two nonstationary detection methods are studied. One is Maximum Likelihood ratio method, while the other combines entropy and the mean alpha angle, which is presented in this dissertation. By using this method, the nonstationary behavior of the azimuthal spectrum can be located and eliminated eventually.
3. The target decomposition and the classification algorithms which are based on polarimetric decomposition are analyzed and simulated, such as Pauli decomposition、Krogager decomposition、H/α、H/A/α、H/α/Wishart、H/A/α/Wishart and Freeman-Wishart classification methods. What is more, a new algorithm, which combines the advantages of Pauli decomposition and AdaBoost algorithm, is proposed in this dissertation. This method can improve the effect of classification and speed up the convergence.
4. Three methods based on subaperture analysis and polarimetric decomposition are researched in this dissertation: the method based on the nonstationary detection and H/αplane, the method combined subaperture decomposition and H/α/Wishart classifier, and the method based on Freeman decomposition and subaperture scattering behavior.
The results of simulating show that the image classification methods, which combine the subaperture data and polarimetric decomposition, can improve the effect of classification and speed up the convergence, and they are valuable tools for practical SAR image classification.
1.总结了目前常用的极化SAR图像分类算法及其存在的问题。在此基础上,将AdaBoost算法与极化通道所携带的信息相结合,对极化SAR图像进行监督分类,在已知场景类别数的情况下,该方法能够加快收敛速度,改善分类效果;
2.研究了基于子孔径分析的极化散射机理与特征提取方法。首先从时频分解的角度对方位向子孔径进行了分析,对场景中存在的各向异性散射行为和布拉格谐振现象进行了讨论。针对场景中的非平稳目标,分析了已有的极大似然比非平稳目标检测算法,还研究了一种基于熵值与平均散射机理角的联合向量检测非平稳目标的方法。通过该方法可以对方位向频谱中的非平稳散射行为进行定位,最终消除非平稳散射在全孔径中的影响;
3.对目标的极化分解方法与基于极化分解的极化SAR图像分类方法进行了研究与仿真试验。如Pauli分解、Krogager分解、H/α分类、H/A/α分类、H/α/Wishart分类、H/A/α/Wishart分类和基于Freeman-Wishart的极化SAR图像分类等。在此基础上,改进了基于AdaBoost算法的极化SAR图像分类方法,将Pauli分解与AdaBoost算法的优势得以发挥,该方法既解决了AdaBoost算法需要知道场景先验知识的缺点,同时还能改善分类效果、提高收敛速度;
4.改进了基于全孔径数据的极化SAR图像分类方法,得到三种基于子孔径的极化SAR图像分类算法并进行了仿真试验:一是非平稳目标检测与H/α分类相结合的极化SAR图像分类;二是子孔径分解与H/α/Wishart迭代分类相结合的极化SAR图像分类;三是结合Freeman分解与子孔径散射特性的极化SAR图像分类。
仿真试验表明,结合子孔径分析与极化分解的极化SAR图像分类,能够在一定程度上改善分类效果、增加分类精细度、提高收敛速度,在实际的SAR图像分类应用中具有很大的价值。
Polarimetric SAR(PolSAR) combines the high space resolution of SAR system and multiple channels data of the targets, which can reveal the physical scattering characteristics better. Therefore, extraction of characteristics and objects classification of PolSAR image play an important part in interpretation of radar data and target recognition. Using polarimetric information fully can express scattering mechanism much more exactly, which can obtain much better result of SAR image. In this dissertation, based on characters of PolSAR data and subaperture analysis, we research on target classification of PolSAR image. The main work and contributions accomplished in the dissertation are as follows:
1. Firstly, the existing popular algorithms of PolSAR image classification and the problems are summarized. Besides, a new classification method, which employs AdaBoost algorithm, is proposed. Having known the types of objects in the scene, the classification accuracy is enhanced by using this method.
2. The subaperture analysis of PolSAR image is studied. Firstly, based on time-frequency analysis, the decomposition theory of subaperture is investigated. Anisotropy polarimetric behavior and Bragg-resonance are also researched, and the latter is one of the important reasons for azimuthal polarimetric changes. Secondly, two nonstationary detection methods are studied. One is Maximum Likelihood ratio method, while the other combines entropy and the mean alpha angle, which is presented in this dissertation. By using this method, the nonstationary behavior of the azimuthal spectrum can be located and eliminated eventually.
3. The target decomposition and the classification algorithms which are based on polarimetric decomposition are analyzed and simulated, such as Pauli decomposition、Krogager decomposition、H/α、H/A/α、H/α/Wishart、H/A/α/Wishart and Freeman-Wishart classification methods. What is more, a new algorithm, which combines the advantages of Pauli decomposition and AdaBoost algorithm, is proposed in this dissertation. This method can improve the effect of classification and speed up the convergence.
4. Three methods based on subaperture analysis and polarimetric decomposition are researched in this dissertation: the method based on the nonstationary detection and H/αplane, the method combined subaperture decomposition and H/α/Wishart classifier, and the method based on Freeman decomposition and subaperture scattering behavior.
The results of simulating show that the image classification methods, which combine the subaperture data and polarimetric decomposition, can improve the effect of classification and speed up the convergence, and they are valuable tools for practical SAR image classification.
引文
[1] Kurpis G P, Booth C J. The New IEEE Standard Dictionary of Electrical and Electronics Terms, 5th ed. New York: IEEE Press, 1993
[2]保铮,邢孟道,王彤.雷达成像技术.北京:电子工业出版社,2005
[3]皮亦鸣,杨建宇.合成孔径雷达成像原理.成都:电子科技大学出版社,2007
[4]庄钊文.雷达极化信息处理与应用.北京:国防工业出版社,1999
[5] Sinclair G. The transmission and reception of elliptically polarized waves. Proceedings of the IRE, 1950, 38(2):148-151
[6]张澄波.综合孔径雷达原理、系统分析与应用.北京:科学出版社,1989
[7] Kennaugh E M. Polarization properties of radar reflections. Antenna Lab., Ohio State University, Columbus, OH, ZRADC cont., No. AF28(009)-90, Project Report. 1952, 389-12(AD2494)
[8] Graves C D. Radar polarization power scattering matrix. Proceedings of the IRE,1956, 44(5): 248-252
[9] Huynen J R. Measurement of the target scattering matrix. Proceedings of the IEEE, 1965, 53(8): 936-946
[10] Huynen J R, Mcnolty F, Hansen E. Component distributions for fluctuating radar targets. IEEE Trans. on Aerospace and Electronic Systems, 1975, 11(6): 1316-1332
[11] Huynen J R. Phenomenological theory of radar targets, Electromagnetic Scattering. P. L. E. Uslerghi. Ed., New York, Academic, 1978
[12] Xi A Q, Boerner W M. The characteristic radar target polarization state theory for the coherent monostatic and reciprocal case using the generalized polarization transformation ratio formulation. AEU, 1990, 44(4): 273-281
[13] Boerner W M. Use of polarization in electromagnetic inverse scattering. Radio Science, 1981, vol.16: 1037-1045
[14] Davidovitz M, Boerner W M. Extension of Kennaugh’s optimal polarization concept to the asymmetric matrix case. IEEE Trans. on Antennas Propagat., 1986, 34(4): 569-574
[15] Van Zyl J J, Burnette C F. Bayesian classification of polarimetric SAR images using adaptive a priori probability. Int. J. Remote Sensing, 1992, 13(5): 835-840
[16] Van Zyl J J. On the importance of polarization in radar scattering problems. Ph. D. thesis, 1986
[17] Tzeng Y C, Chen K S. A fuzzy neural network to SAR image classification. IEEE Trans. on GRS, 1998, 36(1): 301-307
[18] Van Zyl J J. Unsupervised classification of scattering behavior using radar polarimetry data. IEEE Trans. on GRS, 1989, 27(1): 36-45
[19] Cloude S R. Target decomposition theorems in radar scattering. IEE Electronic Letter, 1985, 21(1): 22-24
[20] Cloude S R, Pottier E. A review of target decomposition theorems in radar polarimetry. IEEE Trans. on GRS, 1996, 34(2): 498-518
[21] Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR. IEEE Trans. on GRS, 1997, 35(1): 2249-2259
[22] Pottier E. Unsupervised classification scheme and topography derivation of PolSAR data based on H/A/αpolarimetric decomposition theorem. Proc. 4th Int. Radar Polarimetry, 1998: 535-548
[23] Lee J S, Grunes M R, Ainsworth T L, Du L J, Schuler D L, Clouse S R. Unsupervised classification using polarimetric decomposition and the complex Wishart classifier. IEEE Trans. on GRS, 1999, 37(5): 2249-2258
[24] Yu Yongjian, Huang Shunji, Torre A. Partially correlated k-distribution for multi-look polarimetric Images. CIE International Conference of Radar Proceedings, 1996
[25] Yu Yongjian. Textual-partially correlated polarimetric k-distribution. Proceedings of IGARSS’98, 1998: 2098-2100
[26] Yang Jinhao, Wang Jianguo, Huang Shunji. Speckle filtering for SAR images based on orthonormal wavelet transform. EUSAR, Germany, 1996: 151-154
[27] Liu Guoqing, Huang Shunji, Torre A, Rubertone F. Optimal multi-look polarimetric speckle reduction and its effect on terrain classification. Proceedings of IGARSS’96, 1996, Vol.3: 1571-1573
[28]范立生,高明星,杨健,等.极化SAR遥感中森林特征的提取.电波科学学报,2005,20(5):554-55
[29]刘秀清,杨汝良.基于全极化SAR非监督分类的迭代分类方法.电子学报,2004,31(12):1982-1986
[30]王之禹,朱敏慧,白有天.基于散射模型的极化SAR数据分解.电子与信息学报,2001,23(10):954-961
[31] Ferro-Famil L, Reigber A. Scene characterization using subaperture polarimetric SAR data. IEEE Trans. on GRS, 2003, 41(10): 2264-2276
[32] Ferro-Famil L, Reigber A, Pottier E. Nonstationary natural media analysis from polarimetric SAR data using a two-dimensional time-frequency decomposition approach. Can. J. Remote Sensing, 2005, 31(1): 21-29
[33] Ulaby F T, Elachi C. Radar polarimetry for geosciences applications, Boston: Artech House Inc, 1990
[34] Kostinski A B, Boerner W M. On foundation of radar polarimetry. IEEE Trans. on Antennas Propagat. 1986, 34(12): 1395-1404
[35] Albert G. Mueller and Kennaugh Matrices in Radar Polarimetry. IEEE Trans. on GRS, 1994, 32(3): 590-597
[36]王海江,皮亦鸣,陈红艳.结合ICA相干斑抑制的全极化SAR图像分类.电子学报,2006,34(12):2185-2189
[37] Kong J A, Schwartz A A, Yueh H A. Identification of terrain cover using the optimal polarimetric classifier. J. Electronic Waves and Applications, 1988, 2(2): 171-194
[38] Lee J S, Grunes M R. Classification of multi-look polarimetric SAR imagery based on complex Wishart distribution. Int. J. Remote Sensing, 1994, 15(11): 2299-2311
[39] Pottier E, Saillard J .On radar polarization target decomposition theorems with application to target classification by using network method. Proceedings of ICAP’91, New York, 1991: 265-268
[40] Chen K S, Huang W P, Tsay D H. Classification of mutlifrequency polarimetric SAR imagery using a dynamic learning neural network. IEEE Trans. on GRS, 1996, 34(3): 814-820
[41] Khan K U, Yang J. Novel features for polarimetric SAR images classification by neural network. Proceedings of IGARSS’05, Seoul, Korea, 2005: 165-170
[42] Rignot E, Chellappa R, Dubois P. Unsupervised segmentation of polarimetric SAR data using the covariance matrix. IEEE Trans. on GRS, 1992, 30(4): 697-705
[43] Benz U C. Supervised fuzzy analysis of single- and multi- channel SAR data. IEEE Trans. on GRS, 1999, 37(2): 1023-1037
[44] Pellizzeri T M, Gamba P, Lombardo P. Multitemporal/multiband SAR classification of urban areas using spatial analysis: statistical versus neural kernel-based approach. IEEE Trans. On GRS, 2003, 41(10): 2338-2353
[45] Chen C, Chen K, Lee J. The use of fully polarimetric information for the fuzzy neural classification of SAR images. IEEE Trans. on GRS, 2003, 41(9): 2089-2100
[46] Freund Y, Schapire R. E. A Decision Theoretic Generalization of on-line Learning and an Application to Boosting. Journal of Computer and System Science, 1997, 55(1): 119-139
[47] Boyarshinov V, Magdon-Ismail M. Efficient Optimal Linear Boosting of a Pair of Classifiers. IEEE Trans. on Neural Networks, 2007, 18(2): 317-328
[48] Rignot E, Chellappa R. Segmentation of polarimetric synthetic aperture radar data. IEEE Trans. on Imaging Processing, 1992, 1(3): 281-300
[49] Flake L R, Ahalt S C, Krishnamurthy A K. Detecting anisotropic scattering with hidden Markov models, IEE Proceedings of Radar, 1997, 144(2): 81-86
[50] Souyris J C, Henry C, Adragna F. On the use of complex SAR image spectral analysis for target detection: assessment of polarimetry. IEEE Trans. on GRS, 2003, 41(12):2725-2734
[51] Ferro-Famil, Reigber A, Pottier E, Boerner W-M. Scene characterization using subaperture polarimetric SAR data, IEEE Trans. on GRS, 2003, 41(10): 2264-2276
[52] Ferro-Famil, Reigber A, Pottier E. Nonstationary natural media analysis from polarimetric SAR data using a two-dimensional time-frequency decomposition approach. Can. J. Remote Sensing, 2005, 31(1): 21-29
[53] Moreira A. Real-time synthetic aperture radar (SAR) processing with a new subaperture approach. IEEE Trans. on GRS, 1992, 30(4): 714-722
[54] Goodman N R. Statistical analysis based on a certain multi-variate complex Gaussian distribution(an introduction).Ann. Math. Statist., 1963, 34: 152-177
[55] Lee J S, Grunes M R, Kwok R. Classification of multi-look polarimetric SAR imagery based on the complex Wishart distribution. Int. J. Remote Sensing, 1994, 15(11): 2299-2311
[56] Conradsen K, Nielsen A A, Schou J, Skriver H. Change detection in polarimetric SAR data and the complex Wishart distribution. Proceedings of IGARSS, 2001
[57] Muirhead R J. Aspects of multivariate statistical theory. New York: Wiley, 1982
[58] Cloude S R. Group theory and polarization algebra. Optic, 1986, 75(1):26-36
[59] Krogager E. New decomposition of the radar target scattering matrix. IEE Electronics Letters, 1990, 26(18): 1525-1527
[60] Freeman A, Durden S L. A three-component scattering model for polarimetric SAR data. IEEE Trans. on GRS, 1998, 36(3): 963-973
[61] Lee J S, Grunes M R, Pottier E, Ferro-Famil L. Unsupervised terrain classification preserving polarimetric scattering characteristics. IEEE Trans. on GRS, 2004, 42(4): 722-731
[62] Lee J S, Grunes M R, Grandi G D. Polarimetric SAR speckle filtering and its implication for classification. IEEE Trans. on GRS, 1999, 37(5): 2363-2373
[2]保铮,邢孟道,王彤.雷达成像技术.北京:电子工业出版社,2005
[3]皮亦鸣,杨建宇.合成孔径雷达成像原理.成都:电子科技大学出版社,2007
[4]庄钊文.雷达极化信息处理与应用.北京:国防工业出版社,1999
[5] Sinclair G. The transmission and reception of elliptically polarized waves. Proceedings of the IRE, 1950, 38(2):148-151
[6]张澄波.综合孔径雷达原理、系统分析与应用.北京:科学出版社,1989
[7] Kennaugh E M. Polarization properties of radar reflections. Antenna Lab., Ohio State University, Columbus, OH, ZRADC cont., No. AF28(009)-90, Project Report. 1952, 389-12(AD2494)
[8] Graves C D. Radar polarization power scattering matrix. Proceedings of the IRE,1956, 44(5): 248-252
[9] Huynen J R. Measurement of the target scattering matrix. Proceedings of the IEEE, 1965, 53(8): 936-946
[10] Huynen J R, Mcnolty F, Hansen E. Component distributions for fluctuating radar targets. IEEE Trans. on Aerospace and Electronic Systems, 1975, 11(6): 1316-1332
[11] Huynen J R. Phenomenological theory of radar targets, Electromagnetic Scattering. P. L. E. Uslerghi. Ed., New York, Academic, 1978
[12] Xi A Q, Boerner W M. The characteristic radar target polarization state theory for the coherent monostatic and reciprocal case using the generalized polarization transformation ratio formulation. AEU, 1990, 44(4): 273-281
[13] Boerner W M. Use of polarization in electromagnetic inverse scattering. Radio Science, 1981, vol.16: 1037-1045
[14] Davidovitz M, Boerner W M. Extension of Kennaugh’s optimal polarization concept to the asymmetric matrix case. IEEE Trans. on Antennas Propagat., 1986, 34(4): 569-574
[15] Van Zyl J J, Burnette C F. Bayesian classification of polarimetric SAR images using adaptive a priori probability. Int. J. Remote Sensing, 1992, 13(5): 835-840
[16] Van Zyl J J. On the importance of polarization in radar scattering problems. Ph. D. thesis, 1986
[17] Tzeng Y C, Chen K S. A fuzzy neural network to SAR image classification. IEEE Trans. on GRS, 1998, 36(1): 301-307
[18] Van Zyl J J. Unsupervised classification of scattering behavior using radar polarimetry data. IEEE Trans. on GRS, 1989, 27(1): 36-45
[19] Cloude S R. Target decomposition theorems in radar scattering. IEE Electronic Letter, 1985, 21(1): 22-24
[20] Cloude S R, Pottier E. A review of target decomposition theorems in radar polarimetry. IEEE Trans. on GRS, 1996, 34(2): 498-518
[21] Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR. IEEE Trans. on GRS, 1997, 35(1): 2249-2259
[22] Pottier E. Unsupervised classification scheme and topography derivation of PolSAR data based on H/A/αpolarimetric decomposition theorem. Proc. 4th Int. Radar Polarimetry, 1998: 535-548
[23] Lee J S, Grunes M R, Ainsworth T L, Du L J, Schuler D L, Clouse S R. Unsupervised classification using polarimetric decomposition and the complex Wishart classifier. IEEE Trans. on GRS, 1999, 37(5): 2249-2258
[24] Yu Yongjian, Huang Shunji, Torre A. Partially correlated k-distribution for multi-look polarimetric Images. CIE International Conference of Radar Proceedings, 1996
[25] Yu Yongjian. Textual-partially correlated polarimetric k-distribution. Proceedings of IGARSS’98, 1998: 2098-2100
[26] Yang Jinhao, Wang Jianguo, Huang Shunji. Speckle filtering for SAR images based on orthonormal wavelet transform. EUSAR, Germany, 1996: 151-154
[27] Liu Guoqing, Huang Shunji, Torre A, Rubertone F. Optimal multi-look polarimetric speckle reduction and its effect on terrain classification. Proceedings of IGARSS’96, 1996, Vol.3: 1571-1573
[28]范立生,高明星,杨健,等.极化SAR遥感中森林特征的提取.电波科学学报,2005,20(5):554-55
[29]刘秀清,杨汝良.基于全极化SAR非监督分类的迭代分类方法.电子学报,2004,31(12):1982-1986
[30]王之禹,朱敏慧,白有天.基于散射模型的极化SAR数据分解.电子与信息学报,2001,23(10):954-961
[31] Ferro-Famil L, Reigber A. Scene characterization using subaperture polarimetric SAR data. IEEE Trans. on GRS, 2003, 41(10): 2264-2276
[32] Ferro-Famil L, Reigber A, Pottier E. Nonstationary natural media analysis from polarimetric SAR data using a two-dimensional time-frequency decomposition approach. Can. J. Remote Sensing, 2005, 31(1): 21-29
[33] Ulaby F T, Elachi C. Radar polarimetry for geosciences applications, Boston: Artech House Inc, 1990
[34] Kostinski A B, Boerner W M. On foundation of radar polarimetry. IEEE Trans. on Antennas Propagat. 1986, 34(12): 1395-1404
[35] Albert G. Mueller and Kennaugh Matrices in Radar Polarimetry. IEEE Trans. on GRS, 1994, 32(3): 590-597
[36]王海江,皮亦鸣,陈红艳.结合ICA相干斑抑制的全极化SAR图像分类.电子学报,2006,34(12):2185-2189
[37] Kong J A, Schwartz A A, Yueh H A. Identification of terrain cover using the optimal polarimetric classifier. J. Electronic Waves and Applications, 1988, 2(2): 171-194
[38] Lee J S, Grunes M R. Classification of multi-look polarimetric SAR imagery based on complex Wishart distribution. Int. J. Remote Sensing, 1994, 15(11): 2299-2311
[39] Pottier E, Saillard J .On radar polarization target decomposition theorems with application to target classification by using network method. Proceedings of ICAP’91, New York, 1991: 265-268
[40] Chen K S, Huang W P, Tsay D H. Classification of mutlifrequency polarimetric SAR imagery using a dynamic learning neural network. IEEE Trans. on GRS, 1996, 34(3): 814-820
[41] Khan K U, Yang J. Novel features for polarimetric SAR images classification by neural network. Proceedings of IGARSS’05, Seoul, Korea, 2005: 165-170
[42] Rignot E, Chellappa R, Dubois P. Unsupervised segmentation of polarimetric SAR data using the covariance matrix. IEEE Trans. on GRS, 1992, 30(4): 697-705
[43] Benz U C. Supervised fuzzy analysis of single- and multi- channel SAR data. IEEE Trans. on GRS, 1999, 37(2): 1023-1037
[44] Pellizzeri T M, Gamba P, Lombardo P. Multitemporal/multiband SAR classification of urban areas using spatial analysis: statistical versus neural kernel-based approach. IEEE Trans. On GRS, 2003, 41(10): 2338-2353
[45] Chen C, Chen K, Lee J. The use of fully polarimetric information for the fuzzy neural classification of SAR images. IEEE Trans. on GRS, 2003, 41(9): 2089-2100
[46] Freund Y, Schapire R. E. A Decision Theoretic Generalization of on-line Learning and an Application to Boosting. Journal of Computer and System Science, 1997, 55(1): 119-139
[47] Boyarshinov V, Magdon-Ismail M. Efficient Optimal Linear Boosting of a Pair of Classifiers. IEEE Trans. on Neural Networks, 2007, 18(2): 317-328
[48] Rignot E, Chellappa R. Segmentation of polarimetric synthetic aperture radar data. IEEE Trans. on Imaging Processing, 1992, 1(3): 281-300
[49] Flake L R, Ahalt S C, Krishnamurthy A K. Detecting anisotropic scattering with hidden Markov models, IEE Proceedings of Radar, 1997, 144(2): 81-86
[50] Souyris J C, Henry C, Adragna F. On the use of complex SAR image spectral analysis for target detection: assessment of polarimetry. IEEE Trans. on GRS, 2003, 41(12):2725-2734
[51] Ferro-Famil, Reigber A, Pottier E, Boerner W-M. Scene characterization using subaperture polarimetric SAR data, IEEE Trans. on GRS, 2003, 41(10): 2264-2276
[52] Ferro-Famil, Reigber A, Pottier E. Nonstationary natural media analysis from polarimetric SAR data using a two-dimensional time-frequency decomposition approach. Can. J. Remote Sensing, 2005, 31(1): 21-29
[53] Moreira A. Real-time synthetic aperture radar (SAR) processing with a new subaperture approach. IEEE Trans. on GRS, 1992, 30(4): 714-722
[54] Goodman N R. Statistical analysis based on a certain multi-variate complex Gaussian distribution(an introduction).Ann. Math. Statist., 1963, 34: 152-177
[55] Lee J S, Grunes M R, Kwok R. Classification of multi-look polarimetric SAR imagery based on the complex Wishart distribution. Int. J. Remote Sensing, 1994, 15(11): 2299-2311
[56] Conradsen K, Nielsen A A, Schou J, Skriver H. Change detection in polarimetric SAR data and the complex Wishart distribution. Proceedings of IGARSS, 2001
[57] Muirhead R J. Aspects of multivariate statistical theory. New York: Wiley, 1982
[58] Cloude S R. Group theory and polarization algebra. Optic, 1986, 75(1):26-36
[59] Krogager E. New decomposition of the radar target scattering matrix. IEE Electronics Letters, 1990, 26(18): 1525-1527
[60] Freeman A, Durden S L. A three-component scattering model for polarimetric SAR data. IEEE Trans. on GRS, 1998, 36(3): 963-973
[61] Lee J S, Grunes M R, Pottier E, Ferro-Famil L. Unsupervised terrain classification preserving polarimetric scattering characteristics. IEEE Trans. on GRS, 2004, 42(4): 722-731
[62] Lee J S, Grunes M R, Grandi G D. Polarimetric SAR speckle filtering and its implication for classification. IEEE Trans. on GRS, 1999, 37(5): 2363-2373