一种结合CEM的高光谱遥感影像目标检测算法
|
摘要
提出了一种基于CEM(constrained energy minimum)的高光谱影像目标检测算法。利用虚拟维数实现高光谱影像端元数目的估计,进而获取目标端元。利用基于最大噪声分量算法对原始影像进行降维,在虚拟维数的基础上保留合理的主成分数量。根据提取的目标端元,利用一种改进的CEM算子进行目标检测。实验结果表明:该算法能够获得较高质量的目标检测效果。
A target detection algorithm based on CEM( Constrained Energy Minimum) is proposed.The virtual dimensionality algorithm is employed to effectively estimate the number of endmembers,and the interested endmembers are extracted from hyperspectral images. MNF( Maximum NoiseFraction) is introduced for the dimensionality reduction of original images. Based on the results of virtual dimensionality,the reasonable number of principle component is selected. According to the target endmembers,an improved CEM detector is used for target detection. Experimental results show that the proposed algorithm provides better performance of target detection for hyperspectral images.
A target detection algorithm based on CEM( Constrained Energy Minimum) is proposed.The virtual dimensionality algorithm is employed to effectively estimate the number of endmembers,and the interested endmembers are extracted from hyperspectral images. MNF( Maximum NoiseFraction) is introduced for the dimensionality reduction of original images. Based on the results of virtual dimensionality,the reasonable number of principle component is selected. According to the target endmembers,an improved CEM detector is used for target detection. Experimental results show that the proposed algorithm provides better performance of target detection for hyperspectral images.
引文
[1]REED I S,YU X L.Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution[J].IEEE Transactions on Acoustic,Speech and Signal Processing,1990,38(10):1760-1770.
[2]YU X L,REED I S,STOCKER A D.Comparative performance analysis of adaptive multispectral detectors[J].IEEE Transactions on Signal Processing,1993,41(8):2639-2656.
[3]HARSANYI J C.Detection and classification of subpixel spectral signatures in hyperspectral image sequences[D].Maryland:University of Maryland,1993.
[4]耿修瑞,赵永超.高光谱遥感图像小目标探测的基本原理[J].中国科学,2007,37(8):1081-1087.
[5]张文希,郑茂,李纲.基于端元提取的超光谱图像目标检测算法[J].电光与控制,2010,17(8):45-48.
[6]孙康,耿修瑞,唐海蓉,等.一种基于非线性主成分分析的高光谱图像目标检测方法[J].测绘通报,2015(1):105-108.
[7]粘永健,张志,王力宝,等.基于Fast ICA的高光谱图像目标分割[J].光子学报,2010,39(6):1003-1009.
[8]赵辽英,沈银河,厉小润,等.基于数学形态学的高光谱图像组合核目标检测[J].光学学报,2011,31(12):1-6.
[9]凌强,黄树彩,韦道知,等.联合表示求解二元假设模型的高光谱目标检测[J].电子学报,2016,44(11):2633-2638.
[10]廖佳俊,刘志刚,姜江军,等.基于稀疏表示分步重构算法的高光谱目标检测[J].红外技术,2016,38(8):699-704.
[11]邓贤明,苗放,翟涌光,等.基于形态学的两种高光谱目标探测改进算法[J].中山大学学报(自然科学版),2017,56(1):151-160.
[12]刘翔,张晓杰,郑翰,等.复杂背景中红外多光谱目标检测算法研究[J].上海航天,2016,33(4):56-62.
[13]CHANG C I,DU Q.Estimation of number of spectrally distinct signal sources in hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):608-619.
[14]吴波,张良培,李平湘.非监督正交子空间投影的高光谱混合像元自动分解[J].中国影像图形学报,2004,9(11):1392-1396.
[15]耿修瑞.高光谱遥感图像目标检测与分类技术研究[D].北京:中国科学院遥感应用研究所,2005.
[16]DU Q,SZU H.Interference and noise adjusted principal components analysis for hyperspectral remote sensing image compression[C]//Proceedings of SPIE.Orlando,USA:[s.n.],2006.
[17]苏令华,李纲,衣同胜,等.一种稳健的高光谱图像压缩方法[J].光学精密工程,2007,15(10):1609-1613.
[18]RAMAKRISHNA B,WANG J,CHANG C I,et al.Spectral/spatial hyperspectral image compression in conjunction with virtual dimensionality[C]//Proceedings of the SPIE.Boston,USA:[s.n.],2005,5806:772-781.
[19]粘永健,王展,万建伟.面向异常检测的高光谱图像压缩技术[J].国防科技大学学报,2009,31(3):48-52.
[2]YU X L,REED I S,STOCKER A D.Comparative performance analysis of adaptive multispectral detectors[J].IEEE Transactions on Signal Processing,1993,41(8):2639-2656.
[3]HARSANYI J C.Detection and classification of subpixel spectral signatures in hyperspectral image sequences[D].Maryland:University of Maryland,1993.
[4]耿修瑞,赵永超.高光谱遥感图像小目标探测的基本原理[J].中国科学,2007,37(8):1081-1087.
[5]张文希,郑茂,李纲.基于端元提取的超光谱图像目标检测算法[J].电光与控制,2010,17(8):45-48.
[6]孙康,耿修瑞,唐海蓉,等.一种基于非线性主成分分析的高光谱图像目标检测方法[J].测绘通报,2015(1):105-108.
[7]粘永健,张志,王力宝,等.基于Fast ICA的高光谱图像目标分割[J].光子学报,2010,39(6):1003-1009.
[8]赵辽英,沈银河,厉小润,等.基于数学形态学的高光谱图像组合核目标检测[J].光学学报,2011,31(12):1-6.
[9]凌强,黄树彩,韦道知,等.联合表示求解二元假设模型的高光谱目标检测[J].电子学报,2016,44(11):2633-2638.
[10]廖佳俊,刘志刚,姜江军,等.基于稀疏表示分步重构算法的高光谱目标检测[J].红外技术,2016,38(8):699-704.
[11]邓贤明,苗放,翟涌光,等.基于形态学的两种高光谱目标探测改进算法[J].中山大学学报(自然科学版),2017,56(1):151-160.
[12]刘翔,张晓杰,郑翰,等.复杂背景中红外多光谱目标检测算法研究[J].上海航天,2016,33(4):56-62.
[13]CHANG C I,DU Q.Estimation of number of spectrally distinct signal sources in hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):608-619.
[14]吴波,张良培,李平湘.非监督正交子空间投影的高光谱混合像元自动分解[J].中国影像图形学报,2004,9(11):1392-1396.
[15]耿修瑞.高光谱遥感图像目标检测与分类技术研究[D].北京:中国科学院遥感应用研究所,2005.
[16]DU Q,SZU H.Interference and noise adjusted principal components analysis for hyperspectral remote sensing image compression[C]//Proceedings of SPIE.Orlando,USA:[s.n.],2006.
[17]苏令华,李纲,衣同胜,等.一种稳健的高光谱图像压缩方法[J].光学精密工程,2007,15(10):1609-1613.
[18]RAMAKRISHNA B,WANG J,CHANG C I,et al.Spectral/spatial hyperspectral image compression in conjunction with virtual dimensionality[C]//Proceedings of the SPIE.Boston,USA:[s.n.],2005,5806:772-781.
[19]粘永健,王展,万建伟.面向异常检测的高光谱图像压缩技术[J].国防科技大学学报,2009,31(3):48-52.