Astrophysical image separation by blind time--frequency source separation methods

Authors:
Mehmet Tankut Özgen;Ercan Engin Kuruoğlu;Diego Herranz
Affiliations:
Department of Electrical and Electronics Engineering, Anadolu University, 26470 Eskişehir, Turkey;Instituto di Scienza e Tecnologie dell'Informazione, CNR, via G. Moruzzi 1, 56124 Pisa, Italy;Instituto de Fisica de Cantabria, CSIC-UC, Av. los Castros s/n 39005, Santander, Cantabria, Spain
Venue:
Digital Signal Processing
Year:
2009

Citing 9
Cited 1

A fast fixed-point algorithm for independent component analysis

Neural Computation
Source separation in astrophysical maps using independent factor analysis

Neural Networks - 2003 Special issue: Neural network analysis of complex scientific data: Astronomy and geosciences
Image separation using particle filters

Digital Signal Processing
A Comparative Study of Approximate Joint Diagonalization Algorithms for Blind Source Separation in Presence of Additive Noise

IEEE Transactions on Signal Processing
Nonorthogonal Joint Diagonalization/Zero Diagonalization for Source Separation Based on Time-Frequency Distributions

IEEE Transactions on Signal Processing
Nonorthogonal Joint Diagonalization Free of Degenerate Solution

IEEE Transactions on Signal Processing
Non-orthogonal joint diagonalization in the least-squares sensewith application in blind source separation

IEEE Transactions on Signal Processing
Quadratic optimization for simultaneous matrix diagonalization

IEEE Transactions on Signal Processing
Blind source separation based on time-frequency signalrepresentations

IEEE Transactions on Signal Processing

Blind source separation algorithm based on PSO and algebraic equations of order two

AICI'11 Proceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part III

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, two prevalent blind time-frequency (TF) source separation methods in the literature are adapted to astrophysical image mixtures and four algorithms are developed to separate them into their astrophysical components. The components considered in this work are cosmic microwave background (CMB) radiation, galactic dust and synchrotron, among which the CMB component is emphasized. These simulated components mixed via realistic coefficients are subjected to simulated additive, nonstationary Gaussian noise components of realistic power levels, to yield image mixtures on which our orthogonal and nonorthogonal TF algorithms are applied. The developed algorithms are compared with the FastICA algorithm and CMB component is found to be recovered with an improvement reaching to 3.25 decibels from CMB-synchrotron mixtures. The proposed techniques are believed to be generically applicable in separating other types of astrophysical components as well.