Considering a lot of mixed image pixels are contained in remote sensing images, weighted posterior probability support vector machines are introduced to deal with the remote sensing images unmixing. Weighted posterior probability support vector machines aremotivated by statistical learning theory and is the further research result of regular SVM. Each ground posterior probability can be computed when training samples are determined by the proposed method. Non-linear unmixing mixed pixels model precision are improved by the method because weighted posterior probability support vector machines can avoid the effect on classifier than SVM. In order to solve the multi-class problem, two-class classification methods has been extended to multi-class classification methods and many algorithms have been developed. There are three classes can be described as following, one to many combination model, one to one combination model, and SVM decision tree. With weighted posterior probability support vector machines used on sub-pixel unmixing on remote sensing images, the classifier number are depressed than remote sensing images unmixing without weighted posterior probability support vector machines. The classification result based on weighted posterior probability support vector machines are more accurate according to empirical knowledge. Sample weighting is the main reason avoiding the negative effect of ambiguous ground class. The ground object endmembers can be determined by the proposed method and the posterior probability also be count out at the same time. The result of posterior probability is considered as the percent of each ground object belong to a pixel of remote sensing images. Multi-channel remote images data are used to validate the proposed method in this paper. The experiment results show that the unmixing model based on weighted posterior probability support vector machines has been improved over support vector machines algorithms. The precision of unmixing result obtained based on the proposed method is better than those of support vector machines algorithms.
HU Han, SUN Yong-Hua, LI Xiao-Juan
. Unmixing of Remote Sensing Images Based onWeighted Posterior Probability Support Vector Machines[J]. Journal of Geo-information Science, 2013
, 15(2)
: 249
-254
.
DOI: 10.3724/SP.J.1047.2013.00249
[1] 陈述彭,童庆禧,郭华东.遥感信息机理研究[M].北京:科学出版社,1998, 364-385.
[2] 张彦,邵美珍.基于径向基函数神经网络的混合像元分解[J].遥感学报,2002,4(4):285-289.
[3] Liu W G, Wu E Y, Gopal S, et al.. ARTMMAP: A neuralnetwork approach to subpixel classification[J]. IEEE Transactionson Geoscience and Remote Sensing, 2004, 42(9):1976-1983.
[4] Chang C I, Chiang S S, Smith J A. Linear spectral randommixture analysis for hyperspectral imagery[J]. IEEE Transactionson Geosciences and Remote Sensing, 2002,40(2):375-392.
[5] Huang C, Townshend J. A stepwise regression tree for nonlinearapproximation: Applications to estimating subpixelland cover[J]. International Journal of Remote Sensing,2003,24 (1):75-90.
[6] 唐世浩,朱启疆,闫广建,等.遗传算法及其在遥感线性、非线性模型反演中的应用效果分析[J].北京师范大学学报(自然科学版),2002,38(2):266-272.
[7] Vapnik V. The nature of statistical learning theory [M].New York: Springer Verlag, 1995.
[8] Foody G M, Mathur A. A relative evaluation of multiclassimage classification by support vector machines[J]. IEEETransactions on Geoscience and Remote Sensing, 2004, 42(6): 1335-1343.
[9] 骆剑承,周成虎,梁怡.支持向量机及其遥感影像空间特征提取和分类的应用研究[J].遥感学报,2002,6(1):50-55.
[10] 张锦水,何春阳,潘耀忠,等.基于SVM的多源信息复合的高空间分辨率遥感数据分类研究[J].遥感学报,2006,10(1):49-56.
[11] Weston J, Watkins C. Support vector machines formulti-class pattern recognition[C]. // Proc. of Seventh EuropeanSymposium on Artificial Neural Networks. Bruges:D-Facto Press,1999,219-224.
[12] Li Z Y, Tang S W. Face recognition using improved pairwisecoupling support vector machines[C]. // Proc. of theNinth International Conference on Neural InformationProcessing. Cambridge: MIT Press, 2002,876-880.
[13] Melgani F, Bruzzone L. Classification of hyperspectral remotesensing images with support vector machines[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(8):1778-1790.
[14] Foody G M. Relating the land cover composition ofmixed pixels of artificial neural network classificationoutput[J]. Photogrammtry Engineering and Remote Sensing,1996, 62(5):491-499.
[15] Eric D. On the use of prior and posterior information inthe subpixel proportion problem[J]. IEEE Transactions onGeoscience and Remote Sensing, 2003,41(11):2687-2691.
[16] Valls G C, Chova L G, Maravilla J C, et al. Robust supportvector method for hyperspectral data classificationand knowledge discovery[J]. IEEE Transactions on Geoscienceand Remote Sensing, 2004,42 (7):1530-1542.
[17] Tao Q, Wu G, Wang F, et al.. Posterior probability supportvector machines for unbalanced data[J]. IEEE Transon Neural Networks, 2005,16(6):1561-1573.
[18] 李晶皎译.模式识别(第3 版)[M].北京:电子工业出版社,2006.
[19] Fumitake Takahashi, Shigeo Abe. Decision-tree-basedmulti-class support vector machines[C]. Proceedings ofthe 9th International Conference on Neural Information,Singapore, 2002:1418-1422.
[20] Wu T F, Lin C J, Weng R C. Probability est imates formulti-class classification by pairwise coupling [J]. Journalof Machine Learning Research, 2004,5(2):975-1005.
[21] 李慧,王云鹏,李岩,等.基于SVM和PWC的遥感影像混合像元分解[J].测绘学报,2009,38(4):318-323.
[22] 张翔,肖小玲.支持向量机方法中加权后验概率建模方法[J].清华大学学报:信息科学版,2007,47(10):34-40.
[23] 许菡,李小娟. Normalized Cut 与分水岭变换在高光谱影像混合像元端元提取中的应用[J].中国图象图形学报,2012,17(7):880-885.
[24] Platt J C. Probabilistic outputs for support vector machinesand comparisons to regularized likelihood methods[M]. In: Advances in Large Margin Classifies. Cambridge,MA:MIT Press.1999:70-83.
[25] 吴波,张良培,李平湘.基于支撑向量机概率输出的高光谱影像混合像元分解[J].武汉大学学报:信息科学版,2006,31(1):52-54.
