Independent component analysis, a new concept?
Signal Processing - Special issue on higher order statistics
Jacobi Angles for Simultaneous Diagonalization
SIAM Journal on Matrix Analysis and Applications
A fast fixed-point algorithm for independent component analysis
Neural Computation
High-order contrasts for independent component analysis
Neural Computation
Equivariant adaptive source separation
IEEE Transactions on Signal Processing
Statistical pattern recognition in remote sensing
Pattern Recognition
Hi-index | 0.00 |
In this paper, a joint cumulant independent component analysis (JC-ICA) algorithm is presented. It utilizes the higher order joint cumulants to extract independent components and can be implemented efficiently by a neural network. Its application in SAR (synthetic aperture radar) image analysis is presented and a comparison is also made with two other ICA methods. The results show the usage in image analysis and separation. Because the algorithm is based on statistics of order higher than the second, it is suitable also for applications to data with non-Gaussian distributions in blind signal processing.