基于多核支持向量机的高光谱影像非线性混合像元分解

谭熊; 余旭初; 张鹏强; 秦进春

doi:10.3788/OPE.20142207.1912

您当前的位置：

首页 >

文章列表页 >

基于多核支持向量机的高光谱影像非线性混合像元分解

信息科学 | 更新时间：2020-08-13

- 基于多核支持向量机的高光谱影像非线性混合像元分解
- Nonlinear mixed pixel decomposition of hyperspectral imagery based on multiple kernel SVM
- 光学精密工程 2014年22卷第7期页码：1912-1920
- 作者机构：
  
  1. 信息工程大学,河南郑州,450002
  2. 东华理工大学江西省数字国土重点实验室,江西南昌,330000
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目（No.41201477）();江西省数字国土重点实验室开放基金资助项目（No.DLLJ201403）()
- DOI：10.3788/OPE.20142207.1912
  中图分类号： TP751
- 收稿日期：2013-10-08，
  
  修回日期：2013-12-02，
  
  纸质出版日期：2014-07-25
- 稿件说明：
移动端阅览
谭熊, 余旭初, 张鹏强等. 基于多核支持向量机的高光谱影像非线性混合像元分解[J]. 光学精密工程, 2014,22(7): 1912-1920

TAN Xiong, YU Xu-chu, ZHANG Peng-qiang etc. Nonlinear mixed pixel decomposition of hyperspectral imagery based on multiple kernel SVM[J]. Editorial Office of Optics and Precision Engineering, 2014,22(7): 1912-1920
谭熊, 余旭初, 张鹏强等. 基于多核支持向量机的高光谱影像非线性混合像元分解[J]. 光学精密工程, 2014,22(7): 1912-1920 DOI： 10.3788/OPE.20142207.1912.

TAN Xiong, YU Xu-chu, ZHANG Peng-qiang etc. Nonlinear mixed pixel decomposition of hyperspectral imagery based on multiple kernel SVM[J]. Editorial Office of Optics and Precision Engineering, 2014,22(7): 1912-1920 DOI： 10.3788/OPE.20142207.1912.

摘要

针对基于线性模型分解高光谱影像混合像元分解精度低，而非线性模型难以建立等问题，提出了利用多核支持向量机（MKSVM）的后验概率进行高光谱影像非线性混合像元分解的方法。该方法在支持向量机的基础上，以线性加权组合核函数代替单核函数，采用简单多核学习方法迭代解算权系数来实现分类。然后，通过S型函数将分类器输出值转化为概率；将两两配对概率转换为多类后验概率。最后，利用后验概率实现高光谱影像的非线性混合像元分解。采用该方法对两组推帚式超光谱成像仪（PHI）的高光谱影像进行了对比实验，结果表明：该方法的分类精度分别提高到95.62%和91.51%，均方根误差（RMSE）最小分别为11.15%和7.55%，均小于15%。实验结果显示提出的方法基本消除了混合像元对高光谱影像分类的影响，提高了分类精度。

Abstract

As the mixed pixel decomposition based on linear spectrum models has lower decomposition accuracy and the nonlinear spectrum model is difficult to be established

a nonlinear mixed pixel decomposition method for the hyperspectral imagery was proposed based on the posterior probability of Multiple Kernel Support Vector Machine (MKSVM).On the basis of the SVM

the multiple kernel function formed by linear weighted combination was taken to replace the single kernel and the simple multiple kernel learning was used to solve the weights iteratively to achieve the classification.Then

the output values of the classifier were converted to pairwise coupling probabilities by the sigmoid function and then to the multi-class posterior probability.Finally

the hyperspectral imagery decomposition was achieved through the posterior probability.The results from experiments of two push-broom Hyperspectral Imagers (PHIs) show that the classification accuracies of hyperspectral imagery nonlinear mixed pixel decomposition based on MKSVM reach 95.62% and 91.51%

respectively

the Root Mean Square Errors(RMSEs) are reduced to 11.15% and 7.55%

and both are less than 15%.In conclusion

the influence of mixed pixel on hyperspectral imagery classification is eliminated

and the classification accuracy is increased.

关键词

Keywords

references

黄鸿,杨媚,张满菊. 基于稀疏鉴别嵌入的高光谱遥感影像分类[J]. 光学精密工程,2013,21(11): 2922-2930. HUANG H,YANG M,ZHANG M J. Hyperspectral remote sensing image classification based on SDE[J]. Opt. Precision Eng. ,2013,21(11): 2922-2930. (in Chinese)

杜小平,刘明,夏鲁瑞,等. 基于光谱角累加的高光谱图像异常检测算法[J]. 中国光学,2013,6(3): 325-331. DU X P,LIU M,XIA L R,et al. Anomaly detection algorithm for hyperspectral imagery based on summation of spectral angles[J]. Chinese Journal of Optics,2013,6(3):325-331. (in Chinese)

FOODY G M. Relating the land cover composition of mixed pixels of artificial neural network classification output[J]. Photogrammetry Engineering and Remote Sensing,1996,62(5): 491-499.

KOLACZYK E D. On the use of prior and posterior information in the subpixel proportion problem[J]. IEEE Transactions on Geoscience and Remote Sensing,2003,41(11): 2687- 2691.

吴波,张良培,李平湘. 基于支撑向量回归的高光谱影像混合像元非线性分解[J]. 遥感学报,2006,10(3):312-318. WU B,ZHANG L P,LI P X. Unmixing hyperspectral imagery based on support vector nonlinear approximating regression[J]. Journal of Remote Sensing,2006,10(3):312-318. (in Chinese)

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,2006,36(1):51-54. WU B,ZHANG L P,LI P X. Unmixing of hyperspectral imagery based on probabilistic outputs of support vector machines[J]. Geomatics and Information Science of Wuhan Univesity, 2006,36(1):51-54. (in Chinese)

李惠,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. LI H,WANG Y P,LI Y,et al. Unmixing of remote sensing imagers based on Support Vector Machines and Pairwise Coupling[J]. Acta Geodaetica et Cartographica Sinica,2009,38(4): 318-323. (in Chinese)

杨国鹏,周欣,余旭初,等. 基于相关向量机的高光谱影像混合像元分解[J]. 电子学报,2010,38(12):2751-2756. YANG G P,ZHOU X,YU X CH,et al. Relevance vector machine for hyperspectral imagery unmixing[J]. Acta Electronica Sinica,2010,38(12):2751-2756. (in Chinese)

汪洪桥,孙富春,蔡艳宁,等. 多核学习方法[J]. 自动化学报,2010,36(8):1037-1050. WANG H Q,SUN F CH,CAI Y N,et al. On multiple kernel learning methods[J]. Acta Automatica sinica,2010,36(8):1037-1050. (in Chinese)

RAKOTOMAMONJY A,BACH F,CANU S,et al. Simple MKL[J]. Journal of Machine Learning Research,2008: 1-34.

PLATT J C. Probabilistic outputs for Support Vector Machines and comparisons to regularized likelihood methods. http://research. miscrosoft. com/~jplatt.

WU T F,LIN C J,WENG R C. Probability estimates for multi-class classification by pairwise coupling[J]. Journal of Machine Learning Research,2004,5:975-1005.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据