基于空域滤波的核RX高光谱图像异常检测算法
DOI: 10.369/j.issn.1006-7043.2009.06.020
Keywords: 高光谱 异常检测 非线性 RX 核函数 空域滤波
Abstract:
核RX算法将原始高光谱数据通过非线性映射到高维特征空间进行处理,具有很好的非线性异常检测性能,但当背景数据中混入异常点后背景核矩阵将发生退化,使得漏检率上升.针对此问题,该文提出一种基于空域滤波的核RX算法,利用高光谱图像同一波段相邻像素的空间相关性,采用分波段空域滤波的方式优化背景数据的分布,抑制了异常数据对背景的干扰,构造更加符合背景分布的核矩阵,有效提高检测概率的同时降低了虚警概率.通过试验模拟数据和真实AVIRIS数据的测试检验,说明该算法性能优于传统RX算法以及核RX算法.
| [1] | 1. CHEIN-I C.MINGKAI H Characterization of anomaly detection in hyperspectral imagery 2006(2)
|
| [2] | 5. THORNTON S S.MOURA J M Performance analysis of the adaptive GMRF anomaly detector for hyperspectral imagery 2000
|
| [3] | 6. BASENER B.IENTILUCCI E J.MESSINGER D W Anomaly detection using topology 2007
|
| [4] | ？7. SCHAUM A Spectral subspace matched filtering 2001
|
| [5] | 8. HARSANYI J C.CHANG C I I-lyperspectral image classification and dimensionality reduction:an orthogonal subspace projection approach 1994(4)
|
| [6] | 9. REED I S.YU X Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution 1990(10)
|
| [7] | 10. WEIMIN L.CHEIN-I C A nested spatial window-based approach to target detection for hyperspectral imagery 2004
|
| [8] | 2. CATTERALL S P Anomaly detection based on the statistics of hyperspectral imagery 2004
|
| [9] | ？3. CLARE P E.BERNHARDT M.OXFORD W J A new approach to anomaly detection in hyperspectral images 2003
|
| [10] | 4. HYTLA P.HARDIER C.EISMANN M T Anomaly detection in hyperspectral imagery:A comparison of methods using seasonal data 2007
|
| [11] | 11. HEESUNG K.NASRABADI N M Kernel RX-algorithm:a nonlinear anomaly detector for hyperspectral imagery 2005(2)
|
| [12] | 12. RUIZ A.LOPEZ-DE-TERUEL P E Nonlinear kernelbased statistical pattern analysis 2001(1)
|
| [13] | 13. SCHOLKOPF B.SMOLA A J.MULLER K Nonlinear component analysis as a kernel eigenvalue problem 1998(5)
|
Full-Text
