基于再生核Hilbert空间小波核函数支持向量机的高光谱遥感影像分类.PDFVIP

40 2 Vol.40, No.2 2011 4 Acta Geodaetica et Cartographica Sinica Apr., 2011 2011) Hilbert 谭 琨,杜培军 , 221116 Wavelet Support Vector Machines Based on Reproducing Kernel Hilbert Space for Hyperspectral Remote Sensing Image Classification TAN Kun, DU Peijun Key Laboratory for Terrestrial nvironment and Geohazards Monitoring of State Bureau of Surveying and Mapping, China University of Mining and Technology, Xuzhou 221116, China AbstractSome limitations exist in hyperspectral remote sensing image classification by SVM ( support vector machine) , such as unsatisfactory classification accuracy, difficult kernel parameter selection process a nd depen dence on artificial tricks. In order to solve those problems, the wavelet SVM( WSVM) was proposed based on the investigation to SVMtheory, reproducing kernel Hilbert space( RKHS) and the wavelet analysis. The wavelet kernel in RKHS can approximate arbitrary nonlinea r functions a nd effectively handle the impacts of pa rameter selection. By experimenting the proposed algorithm with hyperspectral image captured by operational modula r imaging spectrometer II (OMIS II) data from China and ROSIS data from Italy. The experimental results shown that Coiflet wavelet kernel functionwas the more effective wavelet interms of classificiatoin accuracy improvement. Key words : hyperspectral remote sensing; wavelet support vector machine ( WSVM) ; reproducing kernel Hilbert space :针对支持向量机用于高光谱遥感影像分类存在的分类精度不高参数 择困难等问题,提出一种再生核 Hilbert 空间的小波核其可以逼近任意非线性函数,能够有效改进参数估计的效果,进而实现基于再生核 Hilbert 空间的小波 核函数支持向量机(小波支持向量机) 并 取北京昌平地区的国产高光谱数据 operationa l modula r imaging spec trometer II ( OMIS II)和意大利 Pavia 大学 ROSIS 高光谱数据进行试验结果表明,应用Coiflet 小波核函数时能获得较 高分类精度 :高光谱遥感; 小波支持向量机; 再生核 Hilbert 空间 :T P751.1 : A : 国家自然科学基金 ( ; 国家 863 计划( 2007AA12Z162) ; 高等学校博士学科点专项科研基金 ( 20070290516) ;江苏省普通高校研究生科研创新计划