New independent component analysis method using higher order statistics with application to remote sensing images

被引：27
作者：
Zhang, XH [1 ]
Chen, CH [1 ]
机构：
[1] Univ Massachusetts, Dept Elect & Comp Engn, N Dartmouth, MA 02747 USA
来源：
OPTICAL ENGINEERING | 2002年 / 41卷 / 07期
关键词：
higher-order statistics; independent component analysis; penalty function; gradient descent; neural networks; remote sensing;
D O I：
10.1117/1.1482722
中图分类号：
O43 [光学];
学科分类号：
070207 ; 0803 ;
摘要：
A new independent component analysis (ICA) method is proposed and a new ICA algorithm making use of the higher order statistics is introduced accordingly. We call it the joint cumulant ICA (JC-ICA) algorithm. Its application in synthetic aperture radar (SAR) and airborne visible infrared imaging spectrometer (AVIRIS) images are discussed. The results show the potential usage in image processing problems. (C) 2002 Society of Photo-Optical Instrumentation Engineers.
引用
收藏
页码：1717 / 1728
页数：12
相关论文
共 22 条
  • [1] Amari S, 1996, ADV NEUR IN, V8, P757
  • [2] AN INFORMATION MAXIMIZATION APPROACH TO BLIND SEPARATION AND BLIND DECONVOLUTION
    BELL, AJ
    SEJNOWSKI, TJ
    [J]. NEURAL COMPUTATION, 1995, 7 (06) : 1129 - 1159
  • [3] Jacobi angles for simultaneous diagonalization
    Cardoso, JF
    Souloumiac, A
    [J]. SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 1996, 17 (01) : 161 - 164
  • [4] Equivariant adaptive source separation
    Cardoso, JF
    Laheld, BH
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1996, 44 (12) : 3017 - 3030
  • [5] Independent component analysis for remote sensing study
    Chen, CH
    Zhang, XH
    [J]. IMAGE AND SIGNAL PROCESSING FOR REMOTE SENSING V, 1999, 3871 : 150 - 158
  • [6] CHEN CH, 2000, PIERS 2000 PROGR ELE
  • [7] CHEN CH, 2000, INT GEOSC REM SENS S, V6, P2620
  • [8] CHIANG SS, 2000, P IGARSS, V7, P3136
  • [9] Modified Herault-Jutten algorithms for blind separation of sources
    Cichocki, A
    Bogner, RE
    Moszczynski, L
    Pope, K
    [J]. DIGITAL SIGNAL PROCESSING, 1997, 7 (02) : 80 - 93
  • [10] INDEPENDENT COMPONENT ANALYSIS, A NEW CONCEPT
    COMON, P
    [J]. SIGNAL PROCESSING, 1994, 36 (03) : 287 - 314
123