有监督的邻域保留嵌入的高光谱遥感影像特征提取算法

测绘学报 ›› 2014, Vol. 43 ›› Issue (5): 508-513.

有监督的邻域保留嵌入的高光谱遥感影像特征提取算法

骆仁波,皮佑国

华南理工大学

收稿日期:2013-12-17 修回日期:2013-09-11 出版日期:2014-05-20 发布日期:2014-06-05
通讯作者: 皮佑国 E-mail:auygpi@scut.edu.cn
基金资助:
广东省教育部产学研结合项目

Supervised Neighborhood Preserving Embedding Feature Extraction of Hyperspectral Imagery

Received:2013-12-17 Revised:2013-09-11 Online:2014-05-20 Published:2014-06-05

摘要/Abstract

摘要：

超光谱遥感图像特征提取对于图像分类具有重要意义，本文提出一种名为判别监督邻域保留嵌入的新型特征提取算法(discriminative supervised neighborhood preserving embedding， DSNPE)。在高维超光谱遥感图像特征提取过程中， DSNPE不但能保留图像的局部流形结构和邻域信息，而且采用像素点由邻域同类像素点线性表示，将邻域中同类和非同类像素点分开处理，利用判别分式求解最优投影矩阵，使高维像素点投影到低维空间时，同类点离得尽可能近，非同类点离得尽可能远，有利于图像的分类。对三幅超光谱遥感图像的特征提取及分类的实验说明：与主成分分析（PCA）、非参数权重特征提取（NWFE）、局部保留投影 (LPP)、邻域保留嵌入（NPE）等相比，具有一定的优越性和可判别性。

关键词: 超光谱遥感图像, 特征提取, 分类

Abstract:

Hyperspectral-image feature extraction is important for image classification. In this paper, A novel hyperspectral remote sensing imagery feature extraction algorithm called discriminative supervised neighborhood preserving embedding (DSNPE) is proposed for supervised linear feature extraction. DSNPE can preserve the local manifold structure and the neighborhood structure. What’s more, for each data point, DSNPE aims at pulling the neighboring points with the same class label towards it as near as possible, while simultaneously pushing the neighboring points with different labels away from it as far as possible. Numerical experiments in three real hyperspectral-image datasets are reported to illustrate the out performance of DSNPE when compare DSNPE with a few competing methods, such as PCA, NWFE, LPP and NPE.

Key words: hyperspectral-image, feature extraction, Classification

中图分类号:

TP751

骆仁波皮佑国. 有监督的邻域保留嵌入的高光谱遥感影像特征提取算法[J]. 测绘学报, 2014, 43(5): 508-513.

参考文献

[1] AGYEMANG T K, HEBLINSKI J. Accuracy Assessment of Supervised Classification of Submersed Macrophytes: the Case of the Gavaraget Region of Lake Sevan, Armenia[J]. Hydrobiologia, 2011,661(1): 85-96 .
[2] TRACOL Y, GUTIERREZ J R. Plant Area Index and Mmicroclimate Underneath Shrub Species from a Chilean Semiarid Community[J]. Journal of Arid Environments, 2011,75(1): 1-6.
[3] TOMLINSON C J, CHAPMAN L. Remote Sensing Land Surface Temperature for Meteorology and Climatology: a Review[J]. Meteorological Applications, 2011,18(3): 296-306.
[4] HUGHES G F. On the mean accuracy of statistical pattern recognizers[J]. IEEE Transactions on Information Theory, 1968,14(1): 55-63.
[5] XIAO Pengfeng, FENG Xuezhi. Segmentation of high-resolutionremotely sensed imagery based on features in frequency domain[ J] .Acta Geodaetica et Cartographica Sinica, 2008, 37( 3): 401. (肖鹏峰，冯学智. 高分辨率遥感图像频域特征提取与图像分割研究[J], 测绘学报，2008,37(3):401).
[6] WU Hangbin. Classification and Feature Extraction of Airborne LiDAR Data Fused with Aerial Image[J]. Acta Geodaetica et Cartographica Sinica,2011,0(1):134.(吴杭彬. 融合航空影像的机载激光扫描数据分类与特征提取[J].测绘学报，2011,40(1):134.)
[7] FONG. M. Dimension reduction on hyperspectral images[R], University of California, Los Angeles, United States, 2007, August 31,.
[8] SCHOTT J. Remote sensing: the image chain approach[M], Oxford University Press, 1996.
[9] FAUVEL M, CHANUSSOT J , BENEDIKTSSON J A. Kernel principal component analysis for the classification of hyperspectral remote-sensing data over urban areas[J],EURASIP Journal on Advances in Signal Processing, 2009, Article ID 783194.
[10] Du Q, Modified Fisher’s linear discriminant analysis for hyperspectral imagery[J], IEEE Geosci. Remote Sens. Lett. 2007, 4(4): 503–507.
[11] BANDOS T V, BRUZZONE L, CAMPS-VALLS G., Classification of hyperspectral images with regularized linear discriminant analysis [J], IEEE Trans. Geosci. Remote Sens., 2009,47 (3):862–873.
[12] YANG J.-M, YY P.-T., KUO B.-C., A nonparametric feature extraction and its application to nearest neighbor classification for hyperspectral image data[J], IEEE Trans. Geosci. Remote Sens., 2010,48(3):1279–1293.
[13] HUANG H.-Y., KUO B.-C., Double nearest proportion feature extraction for hyperspectral-image classi?cation[J], IEEE Trans. Geosci. Remote Sens., 2010, 48(11): 4034–4046.
[14] LUO Renbo,PI Youguo,LIAO Wenzhi. Research on Supervised LPP Feature Extraction for Hyperspectral Image [J].Remote Sensing Technology and Application,2012,27(6):46-52.(骆仁波，皮佑国，廖文志. 超光谱遥感图像有监督LPP特征提取研究[J],遥感技术与应用，2012,27(6):46-52.)
[15] HE X. , CAI D., Yan S., ZhANG H., Neighborhood preserving embedding[C], in Proc. IEEE Int. Conf. Comput. Vis., 2005:1208–1213.
[16] BELKIN M., NIYOGI P., Laplacian eigenmaps and spectral techniques for embedding and clustering[C], in Advances in Neural Infor-mation Processing System, vol. 14. Cambridge, MA: MIT Press, 2002:585–591.
[17] ROWEIS S. T., SAUL L. K., Nonlinear dimensionality reduction by locally linear embedding[J], Science, 2000, 290 (5500): 2323–2326.
[18] AVIRIS NW Indiana’s Indian Pines 1992 Data Set. [Online]. Available: ftp://ftp.ecn.purdue.edu /biehl/MultiSpec/92AV3C(original ?les) and ftp://ftp.ecn.purdue.edu/biehl/PC_MultiSpec/Thy Files.zip (ground truth) .
[19] HAM J., YANGCHI C., CRAWFORD M., GHOSH J., Investigation of the random forest framework for classification of hyperspectral data[J], IEEE Trans. Geosci. Remote Sens., 2005, 43( 3): 492–501.
[20] LANDGREBE D. A., Signal Theory Methods in Multispectral Remote Sensing. Hoboken, NJ: Wiley, 2003.