基于改进双边滤波稀疏表示的高光谱目标检测算法
|
摘要
为了充分利用高光谱图像中包含的空间信息,将一种改进的双边滤波应用到其目标检测中,提出基于光谱角匹配的双边滤波稀疏表示高光谱目标检测算法。通过将光谱角匹配与双边滤波相结合,用高光谱图像像元之间的相似性作为双边滤波器中值域距离的权值,在抑制了图像各波段中噪声的同时突出了目标,然后通过稀疏表示算法进行目标检测。实测的高光谱数据实验显示,与传统稀疏表示方法和普通双边滤波稀疏表示方法比较,所提方法在检测效果上有一定的提高。证明了充分利用高光谱图像的空间信息能进一步提高其目标检测的效果。
In order to make full use of the spatial information contained in the hyperspectral image,an improved bilateral filtering is applied to the target detection,and a bilateral filtering algorithm based on spectral angle matching for sparse representation of hyperspectral target detection is proposed.By combining the spectral angle matching with the bilateral filtering,the similarity between the pixels of hyperspectral image is used as the weight of bilateral filtering.The noise in the band is suppressed and the target is highlighted.Then the target detection is carried out by sparse representation algorithm.Experimental results show that:Compared with the traditional sparse representation method and the sparse representation algorithm with normal bilateral filtering,the proposed method has better detection performance.It is proved that making full use of the spatial information of hyperspectral images can further improve the target detection results.
In order to make full use of the spatial information contained in the hyperspectral image,an improved bilateral filtering is applied to the target detection,and a bilateral filtering algorithm based on spectral angle matching for sparse representation of hyperspectral target detection is proposed.By combining the spectral angle matching with the bilateral filtering,the similarity between the pixels of hyperspectral image is used as the weight of bilateral filtering.The noise in the band is suppressed and the target is highlighted.Then the target detection is carried out by sparse representation algorithm.Experimental results show that:Compared with the traditional sparse representation method and the sparse representation algorithm with normal bilateral filtering,the proposed method has better detection performance.It is proved that making full use of the spatial information of hyperspectral images can further improve the target detection results.
引文
[1]YI C,NASRABADI N M,TRAN T D.Sparse representation for target detection in hyperspectral imagery[J].IEEE Journal of Selected Topics in Signal Processing,2011,5(3):629-640.
[2]YI C,NASRABADI N M,TRAN T D.Simultaneous joint sparsity model for target detection in hyperspectral imagery[J].IEEE Geoscience&Remote Sensing Letters,2011,8(4):676-680.
[3]赵春晖,李晓慧,朱海峰.空间4-邻域稀疏表示的高光谱图像目标检测[J].哈尔滨工程大学学报,2013(9):1171-1178.
[4]FANG L,LI S,KANG X,et al.Spectral-spatial hyperspectral image classification via multiscale adaptive sparse representation[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(12):7738-7749.
[5]ZHANG H,LI J,HUANG Y,et al.A nonlocal weighted joint sparse representation classification method for hyperspectral imagery[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2056-2065.
[6]ZOU J,LI W,DU Q.Sparse representation-based nearest neighbor classifiers for hyperspectral imagery[J].IEEE Geoscience&Remote Sensing Letters,2015,12(12):2418-2422.
[7]SUN W,ZHANG L,ZHANG L,et al.A dissimilarityweighted sparse self-representation method for band selection in hyperspectral imagery classification[J].IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2016,9(9):1-15.
[8]KANG X D,LI S T,BENEDIKTSSON J A.Spectral-spatial hyperspectral image classification with edge-preserving filtering[J].IEEE Transactions on Geoscience&Remote Sensing,2014,52(5):2666-2677.
[9]勾珍珍.基于空间约束和稀疏编码的高光谱图像分类[D].西安:西安电子科技大学,2014.
[10]TOMASI C,MANDUCHI R.Bilateral filtering for gray and color images[C]//International Conference on Computer Vision,1998:839-846.
[11]KRUSE F A,LEFKOFF A B,BOARDMAN J W,et al.The spectral image processing system(SIPS)-interactive visualization and analysis of imaging spectrometer data[J].Remote Sensing of Environment,1993,44(2):145-163.
[12]WANG Y,HUANG S,LIU D,et al.A novel band selection method based on curve area and genetic theory[J].Journal of Optics,2014,43(3):193-202.
[2]YI C,NASRABADI N M,TRAN T D.Simultaneous joint sparsity model for target detection in hyperspectral imagery[J].IEEE Geoscience&Remote Sensing Letters,2011,8(4):676-680.
[3]赵春晖,李晓慧,朱海峰.空间4-邻域稀疏表示的高光谱图像目标检测[J].哈尔滨工程大学学报,2013(9):1171-1178.
[4]FANG L,LI S,KANG X,et al.Spectral-spatial hyperspectral image classification via multiscale adaptive sparse representation[J].IEEE Transactions on Geoscience and Remote Sensing,2014,52(12):7738-7749.
[5]ZHANG H,LI J,HUANG Y,et al.A nonlocal weighted joint sparse representation classification method for hyperspectral imagery[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2014,7(6):2056-2065.
[6]ZOU J,LI W,DU Q.Sparse representation-based nearest neighbor classifiers for hyperspectral imagery[J].IEEE Geoscience&Remote Sensing Letters,2015,12(12):2418-2422.
[7]SUN W,ZHANG L,ZHANG L,et al.A dissimilarityweighted sparse self-representation method for band selection in hyperspectral imagery classification[J].IEEE Journal of Selected Topics in Applied Earth Observations&Remote Sensing,2016,9(9):1-15.
[8]KANG X D,LI S T,BENEDIKTSSON J A.Spectral-spatial hyperspectral image classification with edge-preserving filtering[J].IEEE Transactions on Geoscience&Remote Sensing,2014,52(5):2666-2677.
[9]勾珍珍.基于空间约束和稀疏编码的高光谱图像分类[D].西安:西安电子科技大学,2014.
[10]TOMASI C,MANDUCHI R.Bilateral filtering for gray and color images[C]//International Conference on Computer Vision,1998:839-846.
[11]KRUSE F A,LEFKOFF A B,BOARDMAN J W,et al.The spectral image processing system(SIPS)-interactive visualization and analysis of imaging spectrometer data[J].Remote Sensing of Environment,1993,44(2):145-163.
[12]WANG Y,HUANG S,LIU D,et al.A novel band selection method based on curve area and genetic theory[J].Journal of Optics,2014,43(3):193-202.