Linear Spectral Unmixing of Remote Sensing Images Using Bayesian Method and Non-parametric Model
Received date: 2016-05-28
Revised date: 2016-07-09
Online published: 2016-11-30
To solve the problem of spectral uncertainty in spectral unmixing of remote sensing images, a non-parametric model is proposed to describe the probability distribution of land cover spectrum. A Bayesian method is used to derive a posteriori probability distribution of the proportion of land cover types. Proportions of land cover types are calculated using unbiased estimation and maximum likelihood estimation. Experiment is carried out to map the impervious surface using a Landsat image. The results show that the proposed Bayesian method has higher accuracy than conventional linear spectral unmixing algorithms and the method of hard classification, therefore is effective in solving the spectral uncertainty problem.
Key words： remote sensing; mixed pixel; non-parametric model; end-member; Bayes
LI Xi, CHEN Feng-rui, LI Chang . Linear Spectral Unmixing of Remote Sensing Images Using Bayesian Method and Non-parametric Model[J]. Journal of Applied Sciences, 2016 , 34(6) : 661 -669 . DOI: 10.3969/j.issn.0255-8297.2016.06.002
[1] Heinz D C, Chang C I. Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery [J]. IEEE Transactions on Geoscience and Remote Sensing, 2001, 39(3): 529-545.
[2] Roberts D A, Gardner M, Church R, Ustin S, Scheer G, Green R O. Mapping chaparral in the Santa Monica Mountains using multiple endmember spectral mixture models [J]. Remote Sensing of Environment, 1998, 65(3): 267-279.
[3] Zhu H L. Linear spectral unmixing assisted by probability guided and minimum residual exhaustive search for subpixel classification [J]. International Journal of Remote Sensing, 2005, 26(24): 5585-5601.
[4] Huang C, Townshend J R G. A stepwise regression tree for nonlinear approximation: applications to estimating subpixel land cover [J]. International Journal of Remote Sensing, 2003, 24(1): 75-90.
[5] Huang X, Schneider A, Friedl M A. Mapping sub-pixel urban expansion in China using modis and dmsp/ols nighttime lights [J]. Remote Sensing of Environment, 2016, 175: 92-108.
[6] Liu W, Wu E Y. Comparison of non-linear mixture models: sub-pixel classification [J]. Remote Sensing of Environment, 2005, 94(2): 145-154.
[7] Arai K. Nonlinear mixture model of mixed pixels in remote sensing satellite images based on Monte Carlo simulation [J]. Advances in Space Research, 2008, 41(11), 1715-1723.
[8] Keshava N, Mustard J F. Spectral unmixing [J]. IEEE Signal Processing Magazine, 2002, 19(1): 44-57.
[9] Heinz D, Chang C-I, Althouse M L G. Fully constrained least-squares based linear unmixing [C]//International Geoscience and Remote Sensing Symposium (IGARSS), 1999, 2: 1401-1403.
[10] Roberts D, Halligan K, Dennison P. Viper tools user manual. http://www.vipertools.org/sites/vipertools/files/document/VIPER_Tools_user_manual_v1.5.pdf
[11] Franke J, Roberts D A, Halligan K, Menz G. Hierarchical multiple endmember spectral mixture analysis (MESMA) of hyperspectral imagery for urban environments [J]. Remote Sensing of Environment, 2009, 113 (8): 1712-1723.
[12] 李熙,关泽群,沈体雁. 基于贝叶斯网络的混合像元分解模型[J]. 光电子·激光,2008, 19(8): 1121-1126. Li X, Guan Z Q, Shen T Y. Sub-pixel analysis based on Bayesian network [J]. Journal of Optoelectronics · Laser, 2008, 19(8): 1121-1126. (in Chinese)
[13] 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.
/
| 〈 |
|
〉 |
