针对传统的高光谱数据解混方法中存在的解混精度不高、丰度图模糊的缺陷, 提出一种基于相关向量机的高光谱图像解混方法(unmixing algorithm based on relevance vector machine, UARVM)。其核心思想是采用改进的一对余型的相关向量机将多分类问题转化为多个二分类的问题, 且求取出每个样本所对应的归属类别的概率值, 即丰度值来完成图像的解混。理论研究和仿真结果表明:相对于传统解混方法, UARVM解混精度高, 丰度分布图效果好。
References
[1]
BRAUN A C, WEIDNER U, HINZ S. Classification in high-dimensional feature spaces-assessment using SVM, IVM and RVM with focus on simulated EnMAP data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5(2): 436-443.
[2]
李士进, 常纯, 余宇峰, 等. 基于多分类器组合的高光谱图像波段选择方法[J]. 智能系统学报, 2014, 9(3): 372-378.LI Shijin, CHANG Chun, YU Yufeng, et al. Multi-classifier combination-based hyperspectral band selection[J]. CAAI Transactions on Intelligent Systems, 2014,9(3): 372-378.
[3]
HEINZ D, CHANG C I, ALTHOUSE M L G. Fully constrained least-squares based linear unmixing[J]. IEEE International Geoscience and Remote Sensing Symposium, 1999, 2: 1401-1403.
[4]
XIA Wei, LIU Xuesong, WANG Bin, et al. Independent component analysis for blind unmixing of hyperspectral imagery with additional constraints[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(6): 2165-2179.
[5]
LU Xiaoqiang, WU Hao, YUAN Yuan, et al. Manifold regularized sparse NMF for hyperspectral unmixing[J]. IEEE Transactions on Geoscience and Remote Sensing, 2013, 51(5): 2815-2826.
[6]
DEMIR B, ERTURK S. Improving SVM classification accuracy using a hierarchical approach for hyperspectral Images[C]. Proceedings of 2009 IEEE International Conference on Image Processing (ICIP)., 2009: 2849-2852.
[7]
WANG Liguo, JIA Xiuping. Integration of soft and hard classifications using extended support vector machines [J]. IEEE Geoscience and Remote Sensing Letters, 2009, 6(3): 543-547.
[8]
MIANJI F A, ZHANG Ye. SVM-based unmixing-to-classification conversion for hyperspectral abundance quantification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(11):4318-4327.
[9]
BISHOP C M, TIPPING M E.Variational relevance vector machines[C]//Proceedings of the 16th Conference on Uncertainty in Artificial Intelligence. San Francisco, USA, 2000: 46-53.
[10]
DEMIR B, ERTURK S. Hyperspectral image classifica- tion using relevance vector machines[J].IEEE Geoscience and Remote Sensing Letters, 2007, 4(4): 586-590.
[11]
董超, 赵慧洁. 关联向量机在高光谱影像分类中的应用[J]. 遥感学报, 2010, 14(6): 1273-1284.DONG Chao, ZHAO Huijie. Hyperspectral image classification and application based on relevance vector machine[J]. Journal of Remote Sensing, 2010, 14(6): 1273-1284.
WANG Qunming, WANG Liguo, LIU Danfeng, et al. Sub-pixel mapping for land class with linear features using least square support vector machine[J]. Infrared and Laser Engineering, 2012, 41(6): 1669-1675.
[14]
王立国, 肖倩. 结合Gabor滤波和同质性判定的高光谱图像分类[J]. 应用科技, 2013, 40(4): 21-25.WANG Liguo, XIAO Qian. Hyperspectral imagery classification combined with Gabor filtering and homogeneity discrimination[J]. Applied Science and Technology, 2013, 40(4): 21-25.