基于背景判别与邻域补偿的高光谱异常检测
|
摘要
为克服高光谱局部异常检测算子背景虚警严重,探测效果不佳等问题,提出了基于核光谱角背景判别与邻域补偿的异常检测算法。算法从背景像元的筛选和探测结果的补偿两个角度提高像元探测精度,在背景像元的处理方面,提出了一种基于核光谱角距离相似度的背景像元筛选算法,将光谱分辨性能更强的核光谱角引入背景差异性判别过程,准确可靠地实现局部背景像元的筛选和优化;同时,针对异常检测算子探测精度不高等问题,引入邻域加权的空谱联合补偿机制,并提出基于核光谱角距离相似度的动态模板卷积补偿算法,显著增强了背景与目标的可分性。在与RX、LRX、KRX和CRD等异常检测算法的对比中发现,该算法表现出较强的探测性能,在抑制虚警和提高探测精度等方面达到了不错的效果。
There are serious false alarm and poor detection performance of the local anomaly detection operator for hyperspectral image. To solve these problems,one kind of improved anomaly detection algorithmbased on background discrimination and neighborhood compensation by kernel spectral angle is proposed.The background pixels screening and detecting results compensation are taken into account.In the termof background pixel processing, thealgorithm based on kernel spectral Angle distance similitude is proposed.The kernel spectral angle with stronger spectral resolution is introduced into the background difference discrimination process. In this way,the optimization of local background pixel accurately and reliably is realized.Meanwhile, to solve the problem of low detection accuracy, the joint compensation mechanism of space-spectral characteristics is introduced into neighborhood weighting.The dynamic template convolution compensation algorithm based on kernel spectral Angle distance similarity is proposed, which significantly enhances the separability of background and target. Compared with other abnormal detection algorithms(e.g., RX, LRX, KRX and CRD), the proposed algorithm shows strong detection performance and achieves good effectiveness in suppressing false alarms and improving detection accuracy.
There are serious false alarm and poor detection performance of the local anomaly detection operator for hyperspectral image. To solve these problems,one kind of improved anomaly detection algorithmbased on background discrimination and neighborhood compensation by kernel spectral angle is proposed.The background pixels screening and detecting results compensation are taken into account.In the termof background pixel processing, thealgorithm based on kernel spectral Angle distance similitude is proposed.The kernel spectral angle with stronger spectral resolution is introduced into the background difference discrimination process. In this way,the optimization of local background pixel accurately and reliably is realized.Meanwhile, to solve the problem of low detection accuracy, the joint compensation mechanism of space-spectral characteristics is introduced into neighborhood weighting.The dynamic template convolution compensation algorithm based on kernel spectral Angle distance similarity is proposed, which significantly enhances the separability of background and target. Compared with other abnormal detection algorithms(e.g., RX, LRX, KRX and CRD), the proposed algorithm shows strong detection performance and achieves good effectiveness in suppressing false alarms and improving detection accuracy.
引文
[1] NASRABADI N M.Hyperspectral target detection:An overview of current and future challenges[J].IEEE Signal Processing Magazine,2013,31(1):34- 44.
[2] 高巍,彭宇.基于马氏距离多核学习的高光谱图像分类[J].仪器仪表学报,2018,39(3):250- 257.GAO W,PENG Y.Hyperspectral image classification based on multiplekernel learning with Mahalanobis distance[J].Chinese Journal of Scientific Instrument,2018,39(3):250- 257.
[3] 张兵.高光谱图像处理与信息提取前沿[J].遥感学报,2016,20(5):1062-1090.ZHANG B.Advancement of hyperspectral image processing and information extraction[J].Journal of Remote Sensing,20(5):1062-1090.
[4] MANOLAKIS D,TRUSLOW E,PIEPER M,et al.Detection algorithms in hyperspectral imaging systems:An overview of practical algorithms[J].IEEE Signal Processing Magazine,2014,31(1):24-33.
[5] REED I S,YU X.Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution[J].IEEE Transactions on Acoustics Speech & Signal Processing,1990,38(10):1760-1770.
[6] MOLERO J M,GARZóN E M,GARCíA I,et al.Analysis and optimizations of global and local versions of the RX algorithm for anomaly detection in hyperspectral data[J].IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing,2013,6(2):801- 814.
[7] KWON H,NASRABADI N M.Kernel RX-algorithm:A nonlinear anomaly detector for hyperspectral imagery [J].IEEE Transactions on Geoscience & Remote Sensing,2005,43(2):388-397.
[8] 赵春晖,李晓慧,田明华.采用主成分量化和密度估计期望最大聚类的高光谱异常目标检测[J].光子学报,2013,42(10):1224-1230.ZHAO CH H,LI X H,TIAN M H.Hyperspectral imaging abnormal target detection algorithm using principal component auantization and density estimation on EM clustering[J].Acta PhotonicaSinica,2013,42(10):1024-1230.
[9] 赵春晖,王佳,王玉磊.采用背景抑制和自适应阈值分割的高光谱异常目标检测[J].哈尔滨工程大学学报,2016,37(2):278- 283.ZHAO CH H,WANG J,WANG Y L.Hyperspectral anomaly detection based onbackground suppression and adaptive threshold segmentation[J].Journal of Harbin Engineering University,2016,37(2):278- 283.
[10] 谭海峰,罗天文,杨桄,等.高光谱异常检测中背景抑制方法研究[J].光电子·激光,2016,27(2):177-181.TAN H F,LUO T W,YANG G,et al.Research on background depression in hyperspectral image anomaly detection[J].Journal of Optoelectronics·Laser,2016,27(2):177-181.
[11] ANDREOU C,KARATHANASSI V.A novel multiple endmember spectral mixture analysis using Spectral angle distance[C].Geoscience and Remote Sensing Symposium,IEEE,2012:4110- 4113.
[12] TU B,ZHOU C,KUANG W,et al.Hyperspectral imagery noisy label detection by spectral angle local outlier factor[J].IEEE Geoscience & Remote Sensing Letters,2018,doi:10.1109/LGRS.2018.2842792.
[13] 赵春晖,李佳伟,刘务,等.空谱联合的核光谱角异常检测及GPU实现[J].哈尔滨工程大学学报,2017,38(9):1497-1504.ZHAO CH H,LI J W,LIU W,et al.Anomaly detection and GPU implementation for spatial-spectralkernel spectral angle[J].Journal of Harbin Engineering University,2017,38(9):1497-1504.
[14] VAFADAR M,GHASSEMIAN H.Anomaly detection of hyperspectral imagery using modified collaborative representation[J].IEEE Geoscience & Remote Sensing Letters,2018,15(4):577- 581.
[15] IMANI M.Difference based target detection using mahalanobis distance and spectral angle[J].International Journal of Remote Sensing,2018,40(3):811- 831.
[2] 高巍,彭宇.基于马氏距离多核学习的高光谱图像分类[J].仪器仪表学报,2018,39(3):250- 257.GAO W,PENG Y.Hyperspectral image classification based on multiplekernel learning with Mahalanobis distance[J].Chinese Journal of Scientific Instrument,2018,39(3):250- 257.
[3] 张兵.高光谱图像处理与信息提取前沿[J].遥感学报,2016,20(5):1062-1090.ZHANG B.Advancement of hyperspectral image processing and information extraction[J].Journal of Remote Sensing,20(5):1062-1090.
[4] MANOLAKIS D,TRUSLOW E,PIEPER M,et al.Detection algorithms in hyperspectral imaging systems:An overview of practical algorithms[J].IEEE Signal Processing Magazine,2014,31(1):24-33.
[5] REED I S,YU X.Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution[J].IEEE Transactions on Acoustics Speech & Signal Processing,1990,38(10):1760-1770.
[6] MOLERO J M,GARZóN E M,GARCíA I,et al.Analysis and optimizations of global and local versions of the RX algorithm for anomaly detection in hyperspectral data[J].IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing,2013,6(2):801- 814.
[7] KWON H,NASRABADI N M.Kernel RX-algorithm:A nonlinear anomaly detector for hyperspectral imagery [J].IEEE Transactions on Geoscience & Remote Sensing,2005,43(2):388-397.
[8] 赵春晖,李晓慧,田明华.采用主成分量化和密度估计期望最大聚类的高光谱异常目标检测[J].光子学报,2013,42(10):1224-1230.ZHAO CH H,LI X H,TIAN M H.Hyperspectral imaging abnormal target detection algorithm using principal component auantization and density estimation on EM clustering[J].Acta PhotonicaSinica,2013,42(10):1024-1230.
[9] 赵春晖,王佳,王玉磊.采用背景抑制和自适应阈值分割的高光谱异常目标检测[J].哈尔滨工程大学学报,2016,37(2):278- 283.ZHAO CH H,WANG J,WANG Y L.Hyperspectral anomaly detection based onbackground suppression and adaptive threshold segmentation[J].Journal of Harbin Engineering University,2016,37(2):278- 283.
[10] 谭海峰,罗天文,杨桄,等.高光谱异常检测中背景抑制方法研究[J].光电子·激光,2016,27(2):177-181.TAN H F,LUO T W,YANG G,et al.Research on background depression in hyperspectral image anomaly detection[J].Journal of Optoelectronics·Laser,2016,27(2):177-181.
[11] ANDREOU C,KARATHANASSI V.A novel multiple endmember spectral mixture analysis using Spectral angle distance[C].Geoscience and Remote Sensing Symposium,IEEE,2012:4110- 4113.
[12] TU B,ZHOU C,KUANG W,et al.Hyperspectral imagery noisy label detection by spectral angle local outlier factor[J].IEEE Geoscience & Remote Sensing Letters,2018,doi:10.1109/LGRS.2018.2842792.
[13] 赵春晖,李佳伟,刘务,等.空谱联合的核光谱角异常检测及GPU实现[J].哈尔滨工程大学学报,2017,38(9):1497-1504.ZHAO CH H,LI J W,LIU W,et al.Anomaly detection and GPU implementation for spatial-spectralkernel spectral angle[J].Journal of Harbin Engineering University,2017,38(9):1497-1504.
[14] VAFADAR M,GHASSEMIAN H.Anomaly detection of hyperspectral imagery using modified collaborative representation[J].IEEE Geoscience & Remote Sensing Letters,2018,15(4):577- 581.
[15] IMANI M.Difference based target detection using mahalanobis distance and spectral angle[J].International Journal of Remote Sensing,2018,40(3):811- 831.