组合核函数支持向量机高光谱图像融合分类

高恒振; 万建伟; 粘永健; 王力宝; 徐湛

doi:10.3788/OPE.20111904.0878

您当前的位置：

首页 >

文章列表页 >

组合核函数支持向量机高光谱图像融合分类

信息科学 | 更新时间：2020-08-12

- 组合核函数支持向量机高光谱图像融合分类
- Fusion classification of hyperspectral image by composite kernels support vector machine
- 光学精密工程 2011年19卷第4期页码：878-883
- 作者机构：
  
  国防科技大学电子科学与工程学院,长沙湖南 410073
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(No.40901216)();湖南省研究生科研创新项目(No.CX2010B020)();国防科技大学博士研究生创新基金资助项目(No.B10
- DOI：10.3788/OPE.20111904.0878
  中图分类号： TP751.1
- 收稿日期：2010-10-18，
  
  修回日期：2010-11-23，
  
  网络出版日期：2011-04-26，
  
  纸质出版日期：2011-04-26
- 稿件说明：
移动端阅览
高恒振, 万建伟, 粘永健, 王力宝, 徐湛. 组合核函数支持向量机高光谱图像融合分类[J]. 光学精密工程, 2011,19(4): 878-883

GAO Heng-zhen, WAN Jian-wei, NIAN Yong-jian, WANG Li-bao, XU Zhan. Fusion classification of hyperspectral image by composite kernels support vector machine[J]. Editorial Office of Optics and Precision Engineering, 2011,19(4): 878-883
高恒振, 万建伟, 粘永健, 王力宝, 徐湛. 组合核函数支持向量机高光谱图像融合分类[J]. 光学精密工程, 2011,19(4): 878-883 DOI： 10.3788/OPE.20111904.0878.

GAO Heng-zhen, WAN Jian-wei, NIAN Yong-jian, WANG Li-bao, XU Zhan. Fusion classification of hyperspectral image by composite kernels support vector machine[J]. Editorial Office of Optics and Precision Engineering, 2011,19(4): 878-883 DOI： 10.3788/OPE.20111904.0878.

摘要

针对高光谱图像分类

提出了一种利用组合核函数融合目标光谱域和空域信息的支持向量机学习算法。该算法首先用主成分分析方法对高光谱图像进行特征提取和降维

用虚拟维数估计策略预估原始图像的本征维数

并且在预估的基础上确定要保留的主成份分量数目;然后用数学形态学操作在选取的主分量图像上提取目标的形态信息

得到扩展的空域形态矢量。最后

通过不同的组合策略

构造组合核函数

从而在分类器中引入空域信息

和原有的谱域信息一起

利用支持向量机进行分类。高光谱数据实验表明

在训练时间没有显著差别的情况下

总体分类精度和Kappa系数均提高了2%左右。实验表明

本文提出的方法较单独使用谱域或空域信息进行分类具有一定的优越性。

Abstract

For hyperspectral image classification

a Support Vector Machine (SVM) algorithm with composite kernels was presented to fuse both the spectral information and spatial information of the image. The algorithm adopts Principal Component Analysis (PCA) algorithm to extract the image feature and reduce the dimension for hyperspectral image

and uses the Virtual Dimension (VD) algorithm to estimate the Intrinsic Dimension (ID) of the image. Then

the remained number of Principal Components (PCs) was determined on the basis of the ID.Furthermore

spatial features were extracted by mathematical morphology from the remained PCs

and the Extended Morphological Profile (EMP) vector of image was obtained. By combination of different strategies to construct composite kernels

the spatial information was introduced into the classifier to implement the classification with the SVM and based on both the spectral information and spatial information. Hyperspectral image experiments indicate that the overall accuracy and Kappa coefficients of the proposed approach increase about 2% without increasing the training time obviously. Compared with the classifiers only using the spatial or spectral information

the proposed method shows a lot advantages.

关键词

Keywords

references

陈进. 高光谱图像分类方法研究 . 长沙:国防科学技术大学,2010. CHEN J. On Classification Method of Hyperspectral Images . Changsha: National University of Defense Technology,2010. (in Chinese)[2] MELGANI F,BRUZZONE L. Classification of hyperspectral remote sensing images with support vector machines [J]. IEEE Trans. Geosci. Remote Sens., 2004,42(8):1778-1790.[3] CAMPS-VALLS G, BRUZZONE L. Kernel-based methods for hyperspectral image classification [J]. IEEE Trans. Geosci. Remote Sens., 2005,43(6):1351-1362.[4] MATHIEU F, BENEDIKTSSON J A, CHANUSSOT J, et al.. Spectral and spatial classification of hyperspectral data using SVMs and morphological profiles [J]. IEEE Trans. Geosci. Remote Sens., 2008,46(11):3804-3814.[5] TUIA D, PACIFICI F, KANEVSKI M, et al.. Classification of very high spatial resolution imagery using mathematical morphology and support vector machines [J], IEEE Trans. Geosci. Remote Sens., 2009,47(11):3866-3879.[6] FAUVEL M. Spectral and spatial methods for the classification of urban remote sensing data . Reykjavik:University of Iceland, 2007.[7] TARABALKA Y, BENEDIKTSSON J A, CHANUSSOT J. Spectral-spatial classification of hyperspectral imagery based on partitional clustering techniques [J]. IEEE Trans. Geosci. Remote Sens., 2009,47(8):2973-2987.[8] 粘永健,王展,万建伟,等. 面向异常检测的高光谱图像压缩技术 [J]. 国防科技大学学报,2009,31(3):48-52. NIAN Y J, WANG ZH, WAN J W, et al.. Compression technique for hyperspectral imagery oriented anomaly detection[J]. Journal of National University of Defense Technology,2009,31(3):48-52. (in Chinese)[9] 苏令华,李纲,衣同胜,等. 一种稳健的高光谱图像压缩方法[J]. 光学精密工程,2007,15(10):1609-1615. SU L H, LI G, YI T SH, et al.. A robust hyperspectral image compression method [J]. Opt. Precision Eng., 2007,15(10):1609-1615. (in Chinese)[10] CHANG C I, DU Q. Estimation of number of spectrally distinct signal sources in hyperspectral imagery [J]. IEEE Trans. Geosci. Remote Sens., 2004,42(3):608-619.[11] ftp://ftp.ecn.purdue.edu/biehl/MultiSpec/92 AV3C/ .[12] GONZALEZ R C, WOODS R E. Digital Image Processing [M].USA:Prentice Hall, 2008.

浏览量

992

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

电力巡检中的偏振图像特征融合

基于多尺度空间注意力互补的红外与可见光图像融合

自适应色彩补偿和多尺度融合的水下图像增强

基于直觉模糊集熵测度和显著特征检测的古铜镜X光图像融合

条件扩散和多通道高低频并行的红外与可见光图像融合