收稿：2017-11-30，
录用：2018-4-17，
纸质出版：2019-05
移动端阅览
高光谱图像类内光谱变化较大，“同物异谱”现象普遍存在。利用原始地物光谱特征进行分类精度较低而且分类结果图中存在“椒盐现象”。为了获得好的分类结果，必须充分利用高光谱图像的光谱信息和空间信息，减少类内的光谱变化，并扩大类别间的光谱差异。为此，提出一种滚动引导递归滤波的高光谱图像光谱—空间分类方法。首先，利用主成分分析对高光谱图像进行降维；然后，利用高斯滤波对输入图像进行模糊化，消除图像中的噪声和小尺度结构；接下来，将模糊化后的图像作为引导图像，对输入图像进行边缘保持递归滤波，输出结果作为新的引导图像，重复迭代这个过程直至大尺度边缘被恢复；最后，利用提取的特征波段和支持向量机对高光谱图像进行分类。在两个真实高光谱数据集上进行了分类实验，结果表明本文方法的分类精度优于其他的高光谱图像分类方法。在训练样本极少的情况下，本文方法也能获得较高的分类精度。
Hyperspectral images have relatively large changes in intra-class spectra
and the “same material
different spectra” phenomena are widespread. Moreover
classification accuracy is low when only the original spectral features are considered
and the “pepper and salt phenomenon” can be observed in the classification result map. In the attempt to improve the classification result
the spectral and spatial information of a hyperspectral image needs to be fully utilized to reduce the spectral changes within the class and expand the spectral difference between classes. Thus
in this study
we propose a spectral–spatial hyperspectral image classification method with rolling guidance recursive filtering. First
principal component analysis (PCA) is used to reduce the dimension of the hyperspectral image
in which the main components with large amounts of information are maintained for the input image of the next step. Second
the Gaussian filter is adopted to blur the input image and eliminate the small-scale structure. Third
recursive filtering is performed on the input image to preserve the edge of the image
in which the guidance image is the blurred image
and then the output result is treated as the new guidance image. The process is repeated iteratively until the large-scale edge is restored. Finally
the extracted feature bands of the hyperspectral image are used for classification by a support vector machine. Indian Pines and University of Pavia datasets were initially used to verify the effectiveness of the proposed rolling guidance recursive filtering (RGRF) method. As shown by the homogeneous regions in
Figs. 5
Figs. 5
and
6
6
the noise phenomenon is severe when the extracting feature of PCA is used
whereas the noise is effectively removed when rolling guidance filtering (RGF) and RGRF are utilized. However
at the strong edges
the boundary is blurred and irregular and the spatial smoothness is excessive for RGF. By contrast
the boundary is clear and the strong edges are protected for RGRF. Then
the proposed method is compared with related classification methods by using two real hyperspectral image datasets. The comparative results in
Tables 1
Tables 1
and
2
2
indicate that RGRF can improve classification accuracy more effectively than the other methods. Subsequently
McMemar test was performed to analyze the statistical significance and verify the comparative results. As shown in
Tables 3
Tables 3
and
4
4
the correlations of RGRF with the other methods are statistically significant
and the values exceed
p
>
1.96. The McMemar values in
Tables 3
Tables 3
and
4
4
increase with the decrease in number of samples
which indicates that the advantages of using RGRF becomes more obvious in the case of fewer samples. This study presents an improved edge-preserving filtering method named RGRF. Compared with the other filtering methods
RGRF can maintain the edge structures of land covers and eliminate the texture and noise information within these land covers. RGRF effectively uses the recursive filtering method instead of the bilateral filtering method (i.e.
of the RGF method) in each iteration. In this manner
the continuous spreading of texture and noise information across the strong edges is avoided. Moreover
considering that image quality is significantly improved after RGRF filtering
the classification accuracy of the hyperspectral image is also greatly enhanced. Classification experiments were performed on two real hyperspectral datasets. The results indicate that the classification accuracy of RGRF is superior to the other hyperspectral image classification methods. Moreover
RGRF can achieve higher classification accuracy when only few training samples are used.
Altman E I, Marco G and Varetto F. 1994. Corporate distress diagnosis: comparisons using linear discriminant analysis and neural networks (the Italian experience). Journal of Banking and Finance, 18(3): 505–529
Amjady N, Daraeepour A and Keynia F. 2010. Day-ahead electricity price forecasting by modified relief algorithm and hybrid neural network. IET Generation, Transmission and Distribution, 4(3): 432–444
Balasubramanian M, Schwartz E L, Tenenbaum J B, de Silva V and Langford J C. 2002. The isomap algorithm and topological stability. Science, 295(5552): 7
Baldwin J L, Bateman R M and Wheatley C L. 1990. Application of a neural network to the problem of mineral identification from well logs. The Log Analyst, 31(5): 279–293
Boggs J L, Tsegaye D T, Coleman T L, Reddy K C and Fahsi A. 2003. Relationship between hyperspectral reflectance, soil nitrate-nitrogen, cotton leaf chlorophyll, and cotton yield: a step toward precision agriculture. Journal of Sustainable Agriculture, 22(3): 5–16
Chen C, Zhang L J, Bu J J, Wang C and Chen W. 2010. Constrained Laplacian Eigenmap for dimensionality reduction. Neurocomputing, 73(4/6): 951–958
Cui B G, Ma X D, Xie X Y, Ren G B and Ma Y. 2017. Classification of visible and infrared hyperspectral images based on image segmentation and edge-preserving filtering. Infrared Physics and Technology, 81: 79–88
崔宾阁, 马秀丹, 谢小云. 2017. 小样本的高光谱图像降噪与分类. 遥感学报, 21(5): 728–738
Cui B G, Ma X D and Xie X Y. 2017. Hyperspectral image de-noising and classification with small training samples. Journal of Remote Sensing, 21(5): 728–738 (
Donoho D L and Grimes C. 2003. Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data. Proceedings of the National Academy of Sciences of the United States of America, 100(10): 5591–5596
Fauvel M, Chanussot J and Benediktsson J A. 2009. Kernel principal component analysis for the classification of hyperspectral remote sensing data over urban areas. EURASIP Journal on Advances in Signal Processing, 2009: 783194
Foody G M. 2004. Thematic map comparison. Photogrammetric Engineering and Remote Sensing, 70(5): 627–633
Gastal E S L and Oliveira M M. 2011. Domain transform for edge-aware image and video processing. ACM Transactions on Graphics, 30(4): 69
Gualtieri J A and Cromp R F. 1999. Support vector machines for hyperspectral remote sensing classification//Proceedings of the SPIE 3584, 27th AIPR Workshop: Advances in Computer-Assisted Recognition. Washington, DC, United States: SPIE [DOI: 10.1117/12.339824 http://dx.doi.org/10.1117/12.339824 ]
黄秋燕, 肖鹏峰, 冯学智, 王玥. 2014. 结合相位一致与全变差模型的高分辨率遥感图像边缘检测. 中国图象图形学报, 19(3): 439–446
Huang Q Y, Xiao P F, Feng X Z and Wang Y. 2014. Edge detection from high resolution remote sensing image combining phase congruency with total variation model. Journal of Image and Graphics, 19(3): 439–446 (
Hughes G. 1968. On the mean accuracy of statistical pattern recognizers. IEEE Transactions on Information Theory, 14(1): 55–63
Kang X D, Li S T and Benediktsson J A. 2014a. Feature extraction of hyperspectral images with image fusion and recursive filtering. IEEE Transactions on Geoscience and Remote Sensing, 52(6): 3742–3752
Kang X D, Li S T and Benediktsson J A. 2014b. Spectral–spatial hyperspectral image classification with edge-preserving filtering. IEEE Transactions on Geoscience and Remote Sensing, 52(5): 2666–2677
Li J, Marpu P R, Plaza A, Bioucas-Dias J M and Benediktsson J A. 2013. Generalized composite kernel framework for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing, 51(9): 4816–4829
Lindeberg T. 1994. Scale-space theory: a basic tool for analyzing structures at different scales. Journal of Applied Statistics, 21(1/2): 225–270
Palumbo N F, Harrison G A, Blauwkamp R A and Marquart J K. 2010. Guidance filter fundamentals. Johns Hopkins Apl Technical Digest, 29(1): 60–70
Paris S, Kornprobst P and Tumblin J. 2009. Bilateral Filtering. Hanover, MA, USA: Now Publishers Inc.: 1911–1938
Peng H Y, Jiang C F, Fang X and Liu J S. 2014. Variable selection for Fisher linear discriminant analysis using the modified sequential backward selection algorithm for the microarray data. Applied Mathematics and Computation, 238: 132–140
Pontius J, Martin M, Plourde L and Hallett R. 2008. Ash decline assessment in emerald ash borer-infested regions: a test of tree-level, hyperspectral technologies. Remote Sensing of Environment, 112(5): 2665–2676
Prasad S and Bruce L M. 2008. Limitations of principal components analysis for hyperspectral target recognition. IEEE Geoscience and Remote Sensing Letters, 5(4): 625–629
Sakar C O, Kursun O and Gurgen F. 2012. A feature selection method based on kernel canonical correlation analysis and the minimum Redundancy-Maximum Relevance filter method. Expert Systems with Applications, 39(3): 3432–3437
Spolaôr N, Cherman E A, Monard M C and Lee H D. 2014. ReliefF for multi-label feature selection//Proceedings of 2013 Brazilian Conference on Intelligent Systems. Fortaleza, Brazil: IEEE [DOI: 10.1109/BRACIS.2013.10 http://dx.doi.org/10.1109/BRACIS.2013.10 ]
van Ruitenbeek F J A, Debba P, van der Meer F D, Cudahy T, van der Meijde M and Hale M. 2006. Mapping white micas and their absorption wavelengths using hyperspectral band ratios. Remote Sensing of Environment, 102(3/4): 211–222
Wang L P. 2015. Feature selection algorithm based on conditional dynamic mutual information. International Journal on Smart Sensing and Intelligent Systems, 8(1): 316–337
Wang Y, Niu R Q and Yu X. 2010. Anisotropic diffusion for hyperspectral imagery enhancement. IEEE Sensors Journal, 10(3): 469–477
Zhang Q, Shen X Y, Xu L and Jia J Y. 2014. Rolling Guidance Filter//Proceedings of the 13th European Conference on Computer Vision. Zurich, Switzerland: Springer: 815–830 [DOI: 10.1007/978-3-319-10578-9_53 http://dx.doi.org/10.1007/978-3-319-10578-9_53 ]
相关作者
相关机构
京公网安备11010802024621
