一种基于核特征空间的鲁棒性高光谱异常探测方法

传统高光谱异常探测器的背景统计信息易受异常目标干扰，鲁棒性较差，且难以探测非线性混合的异常目标.针对此问题，运用核特征投影理论，在异常探测器的背景信息构建中引入鲁棒性分析方法，提出了一种在核特征空间中具有鲁棒性的异常探测方法.该方法可以在不需要确定具体的非线性映射函数下，将高光谱数据从低维空间映射到高维特征空间，背景和目标在特征空间中可以用线性模型表示，并在特征空间中构造鲁棒性的探测器.该方法揭示了地物光谱间的高阶特性，可以较好地反映地物分布复杂的目标光谱特性.通过高光谱真实影像和模拟数据的实验证明：1)本文提出的异常探测方法具有更优的受试者工作特征曲线和曲线下面积统计值，目标和背景的分离度更大；2)在核特征空间内，排除异常目标对背景统计信息的干扰，有助于进一步提高探测准确度；3)特征提取可以更好地利用目标和背景的光谱区分性，是异常探测的重要步骤.

Abstract

Traditional hyperspectral anomaly detectors background statistics are easily contaminated by anomalies and not robust that is difficult to detect anomalies of nonlinear mixed. In response to these problems, the Kernel feature projection theory is utilized, robustness analysis method is introduced in the construction of the anomaly detector background information, and a robust anomaly detection method is proposed. Using this method, c hyperspectral data from lowdimensional space can be mapped to high dimensional feature space without specific nonlinear mapping function, the background and anomaly targets can be expressed by a linear model, and a robust detector is constructed in the feature space. This method reveals highorder features between the objects on ground surface and can reflect complex spectral characteristics of target and surface features distribution. Experiments of real hyperspectral images and simulated data can prove: 1) the proposed anomaly detection method has a better receiver operating characteristic (ROC) curve and area under the curve (AUC) statistics and has a greater degree of separation of the target and background; 2) in kernel feature space, exclusion of anomalies contamination on statistics of the background improve the detection accuracy; 3) feature extraction can make better utilizing spectral diversity distinguish anomalies and background, which is an important step of anomaly detector.

参考文献

[1] STEIN D W J, BEAVEN S G, HOFF L E, et al. Anomaly detection from hyperspectral imagery［J］. Signal Processing Magazine, IEEE, 2002, 19(1): 58-69.

[2] MANOLAKIS D, SHAW G. Detection algorithms for hyperspectral imaging applications［J］. Signal Processing Magazine, IEEE, 2002, 19(1): 29-43.

[3] DU Bo, ZHANG Liangpei. Randomselectionbased anomaly detector for hyperspectral imagery［J］. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(5): 1578-1589.

[4] 杜博，张乐飞，张良培，等. 高光谱图像降维的判别流形学习方法［J］. 光子学报，2013, 42(3): 320-325.

DU Bo, ZHANG Lefei, ZHANG Liangpei, et al. Discriminant manifold learning approach for hyperspectral image dimension reduction［J］. Atca Photonica Sinica, 2013, 42(3): 320-325.

[5] MATTEOLI S, DIANI M, CORSINI G. A tutorial overview of anomaly detection in hyperspectral images［J］. Aerospace and Electronic Systems Magazine, IEEE, 2010, 25(7): 5-28.

[6] REED I S, YU X. Adaptive multipleband CFAR detection of an optical pattern with unknown spectral distribution［J］. IEEE Transactions on Acoustics, Speech and Signal Processing, 1990, 38(10): 1760-1770.

[7] 赵春晖，胡春梅，包玉刚. 一种背景误差累积的高光谱图像异常检测算法［J］. 光子学报，2010, 39(10): 1830-1835.

ZHAO Chunhui, HU Chunmei, BAO Yugang. A kind of anomaly detection algorithm for hyperspectral image through background error cumulation［J］. Atca Photonica Sinica, 2010, 39(10): 1830-1835.

[8] 蒲晓丰，雷武虎，黄涛，等. 基于稳健背景子空间的高光谱图像异常检测［J］. 光子学报，2010, 39(12): 2224-2228.

PU Xiaofeng, LEI Wuhu, HUANG Tao, et al. Robust background subspace based anomaly detection algorithm for hyperspectral imagery［J］. Atca Photonica Sinica, 2010, 39(12): 2224-2228.

[9] BILLOR N, HADI A S, VELLEMAN P F. BACON: blocked adaptive computationally efficient outlier nominators［J］. Computational Statistics & Data Analysis, 2000, 34(3): 279-298.

[10] CARLOTTO M J. A clusterbased approach for detecting manmade objects and changes in imagery［J］. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(2): 374-387.

[11] BANERJEE A, BURLINA P, DIEHL C. A support vector method for anomaly detection in hyperspectral imagery［J］. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(8): 2282-2291.

[12] 王挺，杜博，张良培. 顾及局域信息的核化正交子空间投影目标探测方法［J］. 武汉大学学报信息科学版，2013, 38(2): 200-203.

WANG Ting, DU Bo, ZHANG Liangpei. Kernel orthogonal subspace projection for target detection give consideration to local information［J］. Geomatics and Information Science of Wuhan University, 2013, 38(2): 200-203.

[13] HEESUNG KWON, NASRABADI N M. Kernel RXalgorithm: a nonlinear anomaly detector for hyperspectral imagery［J］. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(2): 388-397.

[14] HEESUNG KWON, NASRABADI N M. Kernel orthogonal subspace projection for hyperspectral signal classification［J］. IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(12): 2952-2962.

[15] HEESUNG KWON, NASRABADI N M. Kernel matched signal detectors for hyperspectral target detection［C］. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005.

[16] HEESUNG KWON, NASRABADI N M. Kernel matched subspace detectors for hyperspectral target detection［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(2): 178-194.

[17] GURRAM P, HEESUNG KWON, HAN T. Sparse kernelbased hyperspectral anomaly detection［J］. Geoscience and Remote Sensing Letters, IEEE, 2012, 9(5): 943-947.

[18] 梅锋，赵春晖，孙岩，等. 基于新型光谱相似度量核的高光谱异常检测算法［J］. 光子学报，2009, 38(12): 3165-3170.

MEI Feng, ZHAO Chunhui, SUN Yan, et al. A novel spectral similarity measurement kernel based anomaly detection method in hyperspectral imagery［J］. Atca Photonica Sinica, 2009, 38(12): 3165-3170.

[19] EDISANTER L, INGRAM J. Hyperspectral anomaly detection based on minimum generalized variance method［C］. SPIE, 2008, 6966: 1-7.

[20] HIRSH R G. A performance charcterization of kernelbased algorithms for anomaly detection in hyperspectral imagery［D］. Maryland: University of Maryland, 2007.

[21] GU Yanfeng, LIU Ying, ZHANG Ye. A selective KPCA algorithm based on highorder statistics for anomaly detection in hyperspectral imagery［J］. Geoscience and Remote Sensing Letters, IEEE, 2008, 5(1): 43-47.

赵锐, 杜博, 张良培. 一种基于核特征空间的鲁棒性高光谱异常探测方法[J]. 光子学报, 2013, 42(8): 883. ZHAO Ruia, DU Bob, ZHANG Liangpeia. An Anomaly Detection Method for Hyperspectral Imagery in Kernel Feature Space Based on Robust Analysis[J]. ACTA PHOTONICA SINICA, 2013, 42(8): 883.

一种基于核特征空间的鲁棒性高光谱异常探测方法

关于本站 Cookie 的使用提示

全站搜索

一种基于核特征空间的鲁棒性高光谱异常探测方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索