SAR影像极化特征的混合高斯模型与分类
|
摘要
针对高分辨率极化合成孔径雷达(SAR)影像中极化特征呈现尖峰拖尾等复杂多样的统计特点,采用混合高斯模型(GMM)对极化特征进行建模,提出了一种约束距离的混合多元高斯分布的参数估计算法。该参数估计算法在贪婪期望最大算法框架下设计约束距离函数,自动估计混合分量的个数和模型参数,进而在贝叶斯框架下实现极化SAR影像的地物分类。对Radarsat-2旧金山等地区三组影像数据的分类结果表明:相比于经典的分类算法,所提GMM分类算法的总体精度提高了7%～10%,且对样本数目的依赖性更小,在城区和耕地区域等异质区域可以得到精度更高的分类结果。
Aiming at the various statistical characteristics such as peak tailing presented in the high-resolution synthetic aperture radar(SAR) images, we model the polarimetric features according to the Gaussian mixture model(GMM) and come up with a constrained distance estimation algorithm for the parameters of multivariate Gaussian mixture distribution. Under the framework of greedy expectation maximum algorithm, a constraint distance function is designed and the number of mixed components and model parameters are automatically estimated in this parameter estimation algorithm. Consequently the classification of polarimetric SAR images is realized under the Bayesian framework. The classification results of three groups of image data from Radarsat-2 in San Francisco and other places indicate that the proposed GMM classification algorithm possesses an overall accuracy higher by 7 %-10 % comparing with those by the classical classification algorithms. Moreover, its dependence on sample number is small. The more accurate classification results can be obtained in heterogeneous regions such as urban and farmland.
Aiming at the various statistical characteristics such as peak tailing presented in the high-resolution synthetic aperture radar(SAR) images, we model the polarimetric features according to the Gaussian mixture model(GMM) and come up with a constrained distance estimation algorithm for the parameters of multivariate Gaussian mixture distribution. Under the framework of greedy expectation maximum algorithm, a constraint distance function is designed and the number of mixed components and model parameters are automatically estimated in this parameter estimation algorithm. Consequently the classification of polarimetric SAR images is realized under the Bayesian framework. The classification results of three groups of image data from Radarsat-2 in San Francisco and other places indicate that the proposed GMM classification algorithm possesses an overall accuracy higher by 7 %-10 % comparing with those by the classical classification algorithms. Moreover, its dependence on sample number is small. The more accurate classification results can be obtained in heterogeneous regions such as urban and farmland.
引文
[1] Goodman J W. Some fundamental properties of speckle[J]. Journal of the Optical Society of America,1976, 66(11):1145-1150.
[2] Deng X P, Lopez-Martinez C, Chen J S, et al.Statistical modeling of polarimetric SAR data:a survey and challenges[J]. Remote Sensing, 2017,9(4):348.
[3] Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR[J].IEEE Transactions on Geoscience and Remote Sensing,1997, 35(1):68-78.
[4] Harant O, Bombrun L, Gay M, et al. Segmentation and classification of polarimetric SAR data based on the KummerU distribution[C]. Proceedings of the Fourth International Workshop on Science and Applications of SAR Polarimetry and Polarimetric Interferometry(POLINSAR), 2009:668.
[5] Fay F A, Clarke J, Peters R S. Weibull distribution applied to sea clutter[C]. IEE Proceedings of Conference on Radar, 1977:101-104.
[6] Trunk G V, George S F. Detection of targets in nonGaussian sea clutter[J]. IEEE Transactions on Aerospace and Electronic Systems, 1970, 6(5):620-628.
[7] Lee J S, Hoppel K W, Mango S A, et al. Intensity and phase statistics of multilook polarimetric and interferometric SAR imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(5):1017-1028.
[8]Lopez-Martinez C, Pottier E, Cloude S R. Statistical assessment of eigenvector-based target decomposition theorems in radar polarimetry[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(9):2058-2074.
[9] Doulgeris A P, Akbari V, Eltoft T. AutomaticPolSAR segmentation with the u-distribution and Markov random fields[C]. 9th European Conference on Synthetic Aperture Radar, 2012:183-186.
[10] She X Q, Qiu X L, Lei B, et al. A classification method based on polarimetric entropy and GEV mixture model for intertidal area of PolSAR image[J]. Journal of Radars, 2017, 6(5):554-563.折小强,仇晓兰,雷斌,等.一种极化熵结合混合GE V模型的全极化SAR潮间带区域地物分类方法[J].雷达学报,2017, 6(5):554-563.
[11] Li H C, Hong W, Wu Y R, et al. On the empiricalstatistical modeling of SAR images with generalized gamma distribution[J]. IEEE Journal of Selected Topics in Signal Processing, 2011, 5(3):386-397.
[12] Gou S P, Qiao X, Zhang X R, et al. Eigenvalue analysis-based approach for POL-SAR image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(2):805-818.
[13] Zhang D M, Fu M S, Guo H, et al. Automatic estimation algorithm of component number of mixture model based on penalized distance[J].Journal of South China University of Technology(Natural Science Edition), 2009, 37(10):101-107.张大明,符茂胜,郭慧,等.基于惩罚距离的混合模型分量数自动估计算法[J].华南理工大学学报(自然科学版),2009, 37(10):101-107.
[14] Akaike H. A new look at the statistical model identification[J]. IEEE Transactions on Automatic Control, 1974, 19(6):716-723.
[15] Wan J H, Zang J X, Liu S W. Fusion and classification of SAR and optical image with consideration of polarization characteristics[J]. Acta Optica Sinica, 2017, 37(6):0628001.万剑华,臧金霞,刘善伟.顾及极化特征的SAR与光学影像融合与分类[J].光学学报,2017, 37(6):0628001.
[16] Xie X F, Xu X, Dong H, et al. A semi-supervised dimension reduction method for polarimetric SAR image classification[J]. Acta Optica Sinica, 2018,38(4):0428001.谢欣芳,徐新,董浩,等.一种极化SAR影像分类中的半监督降维方法[J].光学学报,2018, 38(4):0428001.
[17] Ye Z, Bai L, Nian Y J. Hyperspectral image classification algorithm based on Gabor feature and locality-preserving dimensionality reduction[J].Acta Optica Sinica, 2016, 36(10):1028003.叶珍,白璘,粘永健.基于Gabor特征与局部保护降维的高光谱图像分类算法[J].光学学报,2016,36(10):1028003.
[18] Xie F C. Mixed hidden variable model and greedy EMalgorithm[D]. Nanjing:Southeast University, 2003.解锋昌.混合隐藏变量模型和贪婪EM算法[D].南京:东南大学,2003.
[19] Peng Y J, Chen J Y, Xu X, et al. SAR images statistical modeling and classification based on the mixture of alpha-stable distributions[J]. RemoteSensing, 2013, 5(5):2145-2163.
[20] Dong H, Xu X, Sui H G, et al. Copula-based joint statistical model for polarimetric features and its application in PolSAR image classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2017, 55(10):5777-5789.
[2] Deng X P, Lopez-Martinez C, Chen J S, et al.Statistical modeling of polarimetric SAR data:a survey and challenges[J]. Remote Sensing, 2017,9(4):348.
[3] Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR[J].IEEE Transactions on Geoscience and Remote Sensing,1997, 35(1):68-78.
[4] Harant O, Bombrun L, Gay M, et al. Segmentation and classification of polarimetric SAR data based on the KummerU distribution[C]. Proceedings of the Fourth International Workshop on Science and Applications of SAR Polarimetry and Polarimetric Interferometry(POLINSAR), 2009:668.
[5] Fay F A, Clarke J, Peters R S. Weibull distribution applied to sea clutter[C]. IEE Proceedings of Conference on Radar, 1977:101-104.
[6] Trunk G V, George S F. Detection of targets in nonGaussian sea clutter[J]. IEEE Transactions on Aerospace and Electronic Systems, 1970, 6(5):620-628.
[7] Lee J S, Hoppel K W, Mango S A, et al. Intensity and phase statistics of multilook polarimetric and interferometric SAR imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(5):1017-1028.
[8]Lopez-Martinez C, Pottier E, Cloude S R. Statistical assessment of eigenvector-based target decomposition theorems in radar polarimetry[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(9):2058-2074.
[9] Doulgeris A P, Akbari V, Eltoft T. AutomaticPolSAR segmentation with the u-distribution and Markov random fields[C]. 9th European Conference on Synthetic Aperture Radar, 2012:183-186.
[10] She X Q, Qiu X L, Lei B, et al. A classification method based on polarimetric entropy and GEV mixture model for intertidal area of PolSAR image[J]. Journal of Radars, 2017, 6(5):554-563.折小强,仇晓兰,雷斌,等.一种极化熵结合混合GE V模型的全极化SAR潮间带区域地物分类方法[J].雷达学报,2017, 6(5):554-563.
[11] Li H C, Hong W, Wu Y R, et al. On the empiricalstatistical modeling of SAR images with generalized gamma distribution[J]. IEEE Journal of Selected Topics in Signal Processing, 2011, 5(3):386-397.
[12] Gou S P, Qiao X, Zhang X R, et al. Eigenvalue analysis-based approach for POL-SAR image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(2):805-818.
[13] Zhang D M, Fu M S, Guo H, et al. Automatic estimation algorithm of component number of mixture model based on penalized distance[J].Journal of South China University of Technology(Natural Science Edition), 2009, 37(10):101-107.张大明,符茂胜,郭慧,等.基于惩罚距离的混合模型分量数自动估计算法[J].华南理工大学学报(自然科学版),2009, 37(10):101-107.
[14] Akaike H. A new look at the statistical model identification[J]. IEEE Transactions on Automatic Control, 1974, 19(6):716-723.
[15] Wan J H, Zang J X, Liu S W. Fusion and classification of SAR and optical image with consideration of polarization characteristics[J]. Acta Optica Sinica, 2017, 37(6):0628001.万剑华,臧金霞,刘善伟.顾及极化特征的SAR与光学影像融合与分类[J].光学学报,2017, 37(6):0628001.
[16] Xie X F, Xu X, Dong H, et al. A semi-supervised dimension reduction method for polarimetric SAR image classification[J]. Acta Optica Sinica, 2018,38(4):0428001.谢欣芳,徐新,董浩,等.一种极化SAR影像分类中的半监督降维方法[J].光学学报,2018, 38(4):0428001.
[17] Ye Z, Bai L, Nian Y J. Hyperspectral image classification algorithm based on Gabor feature and locality-preserving dimensionality reduction[J].Acta Optica Sinica, 2016, 36(10):1028003.叶珍,白璘,粘永健.基于Gabor特征与局部保护降维的高光谱图像分类算法[J].光学学报,2016,36(10):1028003.
[18] Xie F C. Mixed hidden variable model and greedy EMalgorithm[D]. Nanjing:Southeast University, 2003.解锋昌.混合隐藏变量模型和贪婪EM算法[D].南京:东南大学,2003.
[19] Peng Y J, Chen J Y, Xu X, et al. SAR images statistical modeling and classification based on the mixture of alpha-stable distributions[J]. RemoteSensing, 2013, 5(5):2145-2163.
[20] Dong H, Xu X, Sui H G, et al. Copula-based joint statistical model for polarimetric features and its application in PolSAR image classification[J]. IEEE Transactions on Geoscience and Remote Sensing,2017, 55(10):5777-5789.