基于黎曼流形的极化SAR图像分类

杨文; 钟能; 严天恒; 杨祥立

doi:10.12000/JR17031

基于黎曼流形的极化SAR图像分类

DOI: 10.12000/JR17031

武汉大学电子信息学院武汉 430072

基金项目: 国家自然科学基金(61331016, 61271401)

详细信息

作者简介:
杨　文(1976–)，男，教授，博士生导师，2004年获得武汉大学工学博士学位。研究方向为图像处理与计算机视觉。E-mail: yangwen@whu.edu.cn

钟　能(1993–)，男，2015年获得吉林大学工学学士学位，现于武汉大学电子信息学院攻读硕士学位。主要研究方向为极化合成孔径雷达图像处理。E-mail: zn_whu@whu.edu.cn

严天恒(1995–)，女，2016年获得武汉大学工学学士学位，现于武汉大学电子信息学院信号处理实验室攻读硕士学位。主要研究方向为极化合成孔径雷达图像解译。E-mail: yanth_eis@whu.edu.cn

杨祥立(1991–)，男，2016年获得武汉大学工学硕士学位，现于武汉大学电子信息学院攻读博士学位。主要研究方向为极化合成孔径雷达图像处理。E-mail: xiangliyang@whu.edu.cn

通讯作者:
杨文 yangwen@whu.edu.cn

中图分类号: TN957
计量
- 文章访问数: 3570
- HTML全文浏览量: 803
- PDF下载量: 767
- 被引次数: 0
出版历程
- 收稿日期: 2017-03-24
- 修回日期: 2017-06-08
- 网络出版日期: 2017-10-28

Classification of Polarimetric SAR Images Based on the Riemannian Manifold

School of Electronic Information, Wuhan University, Wuhan 430072, China

Funds: The National Natural Science Foundation of China (61331016, 61271401)

摘要

摘要: 分类是极化SAR图像解译的核心内容之一。一种新的思路是通过利用极化SAR协方差矩阵所形成的黎曼流形结构特性进行极化SAR图像分类。该文首先回顾了极化SAR图像分析中常用的黎曼流形测度，然后论述了如何对黎曼流形上的极化协方差矩阵进行稀疏编码。在监督分类方面，基于核空间黎曼流形稀疏编码提出了融合空间信息的极化SAR图像监督分类方法；在非监督分类方面，基于黎曼稀疏编码提出了利用黎曼稀疏诱导相似度的极化SAR图像非监督分类方法。在EMISAR和AIRSAR极化数据上的实验结果表明了该文所提方法的有效性。
- 极化SAR /
- 图像分类 /
- 黎曼流形 /
- 稀疏编码
Abstract: Classification is one of the core components in the interpretation of Polarimetric Synthetic Aperture Radar (PolSAR) images. A new PolSAR image classification approach employs the structural properties of the Riemannian manifold formed by PolSAR covariance matrices. In this paper, we first review the Riemannian manifold metrics generally used in PolSAR image analysis. Then, we describe a sparse coding method for the covariance matrices in the Riemannian manifold. For supervised classification, we propose a PolSAR image classification method that considers spatial information based on kernel space sparse coding. As for unsupervised PolSAR image classification, a method that takes advantage of Riemannian sparse induced similarity is proposed. Experimental results on EMISAR and AIRSAR data demonstrate the effectiveness of the proposed methods.
- Polarimetric SAR (PolSAR) /
- Image classification /
- Riemannian manifold /
- Sparse coding

HTML全文

图 1 EMISAR数据的实验结果

Figure 1. The experiment results of EMISAR data

下载: 全尺寸图片幻灯片

图 2 AIRSAR数据的实验结果

Figure 2. The experiment results of AIRSAR data

下载: 全尺寸图片幻灯片

表 1 EMISAR数据监督分类结果

Table 1. The supervised classification results of EMISAR data

方法	针叶林	小麦	油菜	燕麦	黑麦	OA	Kappa
Wishart方法	0.9370	0.9527	0.9466	0.9843	0.9904	0.9713	0.9623
稀疏表达方法	0.9996	0.9860	0.9479	0.8946	0.9621	0.9472	0.9304
KSC方法	1	1	0.9960	0.9982	0.9975	0.9981	0.9975

下载: 导出CSV

表 2 AIRSAR数据非监督分类结果

Table 2. The unsupervised classification results of AIRSAR data

方法	OA	F1-score	Purity	Entropy
Wishart方法	0.6265	0.6084	0.7324	0.2909
Bartlett方法	0.6538	0.6376	0.8015	0.2353
RSC方法	0.8485	0.8633	0.9047	0.1344

下载: 导出CSV

参考文献(32)

[1]	Yang Wen, Song Hui, Xia Gui-song, et al.. Dissimilarity measurements for processing and analyzing PolSAR data: A survey[C]. Proceedings of 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Milan, Italy, 2015: 1562–1565.
[2]	Lee J S, Grunes M R, Ainsworth T L, et al.. Unsupervised classification using polarimetric decomposition and the complex Wishart classifier[J]. IEEE Transactions on Geoscience and Remote Sensing, 1999, 37(5): 2249–2258. doi: 10.1109/36.789621
[3]	Frery A C, Correia A H, and Freitas C D C. Classifying multifrequency fully polarimetric imagery with multiple sources of statistical evidence and contextual information[J]. IEEE Transactions on Geoscience and Remote Sensing, 2007, 45(10): 3098–3109. doi: 10.1109/TGRS.2007.903828
[4]	Anfinsen S N, Jenssen R, and Eltoft T. Spectral clustering of polarimetric SAR data with the Wishart-derived distance measures[C/OL]. Proceedings of the 3rd International Workshop on Science and Applications of SAR Polarimetry and Polarimetric Interferometry, Noordwijk, Netherlands, 2007. http://adsabs.harvard.edu/abs/2007ESASP.644E..10A.
[5]	Banerjee A, Merugu S, Dhillon I S, et al.. Clustering with Bregman divergences[J]. The Journal of Machine Learning Research, 2005, 6: 1705–1749.
[6]	Kersten P R, Lee J S, and Ainsworth T L. Unsupervised classification of polarimetric synthetic aperture radar images using fuzzy clustering and EM clustering[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(3): 519–527. doi: 10.1109/TGRS.2004.842108
[7]	Song H, Yang W, Bai Y, et al.. Unsupervised classification of polarimetric SAR imagery using large-scale spectral clustering with spatial constraints[J]. International Journal of Remote Sensing, 2015, 36(11): 2816–2830. doi: 10.1080/01431161.2015.1043759
[8]	Song Hui, Yang Wen, Xu Xin, et al.. Unsupervised PolSAR imagery classification based on Jensen-Bregman LogDet divergence[C]. Proceedings of the 10th European Conference on Synthetic Aperture Radar, Berlin, Germany, 2014: 1–4.
[9]	Cherian A, Sra S, Banerjee A, et al.. Jensen-Bregman LogDet divergence with application to efficient similarity search for covariance matrices[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(9): 2161–2174. doi: 10.1109/TPAMI.2012.259
[10]	Zhang La-mei, Sun Liang-jie, Zou Bin, et al.. Fully polarimetric SAR image classification via sparse representation and polarimetric features[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(8): 3923–3932. doi: 10.1109/JSTARS.2014.2359459
[11]	Harandi M T, Hartley R, Lovell B, et al.. Sparse coding on symmetric positive definite manifolds using Bregman divergences[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 27(6): 1294–1306. doi: 10.1109/TNNLS.2014.2387383
[12]	Yang Fan, Gao Wei, Xu Bin, et al.. Multi-frequency polarimetric SAR classification based on Riemannian manifold and simultaneous sparse representation[J]. Remote Sensing, 2015, 7(7): 8469–8488. doi: 10.3390/rs70708469
[13]	Song Hui, Yang Wen, Zhong Neng, et al.. Unsupervised classification of PolSAR imagery via kernel sparse subspace clustering[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(10): 1487–1491. doi: 10.1109/LGRS.2016.2593098
[14]	Cherian A and Sra S. Riemannian sparse coding for positive definite matrices[C]. Proceedings of European Conference on Computer Vision (ECCV), Zurich, Switzerland, 2014: 299–314.
[15]	Yang Wen, Zhong Neng, Yang Xiang-li, et al.. Riemannian sparse coding for classification of PolSAR images[C]. Proceedings of 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Beijing, China, 2016: 5698–5701.
[16]	Pennec X, Fillard P, and Ayache N. A Riemannian framework for tensor computing[J]. International Journal of Computer Vision, 2006, 66(1): 41–66. doi: 10.1007/s11263-005-3222-z
[17]	Arsigny V, Fillard P, Pennec X, et al.. Log-Euclidean metrics for fast and simple calculus on diffusion tensors[J]. Magnetic Resonance in Medicine, 2006, 56(2): 411–421. doi: 10.1002/(ISSN)1522-2594
[18]	Jayasumana S, Hartley R, Salzmann M, et al.. Kernel methods on Riemannian manifolds with Gaussian RBF kernels[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(12): 2464–2477. doi: 10.1109/TPAMI.2015.2414422
[19]	Tibshirani R. Regression shrinkage and selection via the lasso: A retrospective[J]. Journal of the Royal Statistical Society, 2011, 73(3): 273–282. doi: 10.1111/rssb.2011.73.issue-3
[20]	Elhamifar E and Vidal R. Sparse subspace clustering: Algorithm, theory, and applications[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(11): 2765–2781. doi: 10.1109/TPAMI.2013.57
[21]	Cherian A and Sra S. Riemannian dictionary learning and sparse coding for positive definite matrices[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016. DOI: 10.1109/TNNLS.2016.2601307.
[22]	Birgin E G, Martínez J M, and Raydan M. Algorithm 813: SPG-software for convex-constrained optimization[J]. ACM Transactions on Mathematical Software, 2001, 27(3): 340–349. doi: 10.1145/502800.502803
[23]	Wright J, Yang A Y, Ganesh A, et al.. Robust face recognition via sparse representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(2): 210–227. doi: 10.1109/TPAMI.2008.79
[24]	Zhang Hong-yan, Li Jia-yi, Huang Yuan-cheng, et al.. A nonlocal weighted joint sparse representation classification method for hyperspectral imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(6): 2056–2065. doi: 10.1109/JSTARS.2013.2264720
[25]	Deledalle C A, Denis L, Poggi G, et al.. Exploiting patch similarity for SAR image processing: The nonlocal paradigm[J]. IEEE Signal Processing Magazine, 2014, 31(4): 69–78. doi: 10.1109/MSP.2014.2311305
[26]	Cheng Hong, Liu Zi-cheng, and Yang Jie. Sparsity induced similarity measure for label propagation[C]. Proceedings of the 2009 IEEE 12th International Conference on Computer Vision, Kyoto, Japan, 2009: 317–324.
[27]	Zelnik-Manor L and Perona P. Self-tuning spectral clustering[C]. Advances in Neural Information Processing Systems, Vancouver, 2004: 1601–1608.
[28]	Achanta R, Shaji A, Smith K, et al.. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11): 2274–2282. doi: 10.1109/TPAMI.2012.120
[29]	Lee J S, Grunes M R, and Kwok R. Classification of multi-look polarimetric SAR imagery based on complex Wishart distribution[J]. International Journal of Remote Sensing, 1994, 15(11): 2299–2311. doi: 10.1080/01431169408954244
[30]	Foody G M. Status of land cover classification accuracy assessment[J]. Remote Sensing of Environment, 2002, 80(1): 185–201. doi: 10.1016/S0034-4257(01)00295-4
[31]	Cherian A, Morellas V, and Papanikolopoulos N. Bayesian nonparametric clustering for positive definite matrices[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(5): 862–874. doi: 10.1109/TPAMI.2015.2456903
[32]	Xu Kan, Yang Wen, Liu Gang, et al.. Unsupervised satellite image classification using Markov field topic model[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(1): 130–134. doi: 10.1109/LGRS.2012.2194770