Mixing matrix identification for underdetermined blind signal separation: using hough transform and fuzzy K-means clustering

Authors:
Tsung-Ying Sun;Ling-Erh Lan;Chan-Cheng Liu;Chih-Li Huo
Affiliations:
Department of Electrical Engineering, National Dong Hwa University, Hualien, Taiwan, R.O.C.;Department of Electrical Engineering, National Dong Hwa University, Hualien, Taiwan, R.O.C.;Institute of Information Science, Academia Sinica, Taipei, Taiwan, R.O.C.;Department of Electrical Engineering, National Dong Hwa University, Hualien, Taiwan, R.O.C.
Venue:
SMC'09 Proceedings of the 2009 IEEE international conference on Systems, Man and Cybernetics
Year:
2009

Citing 9
Cited 0

Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
An Introduction to Digital Image Processing With Matlab

An Introduction to Digital Image Processing With Matlab
Blind source separation using order statistics

Signal Processing
A general solution to blind inverse problems for sparse input signals

Neurocomputing
Blind separation of speech mixtures via time-frequency masking

IEEE Transactions on Signal Processing
Sequential blind extraction of instantaneously mixed sources

IEEE Transactions on Signal Processing
Mutual information approach to blind separation of stationary sources

IEEE Transactions on Information Theory
A Markov model for blind image separation by a mean-field EM algorithm

IEEE Transactions on Image Processing
A spectral clustering approach to underdetermined postnonlinear blind source separation of sparse sources

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper focuses on the underdetermined blind signal separation problem with sparse representation. The algorithm is proposed to identify the parameters of mixing model which are unknown. The distribution of mixtures are mapping to a new histogram domain by Hough transform which converts the Cartesian image space to the normal parameterization. And then, fuzzy k-means clustering is employed to seek the cluster centers, i.e. parameters of mixing model, on the histogram. Obtaining accurate estimates, the sources can be recovered clearly. The proposed algorithm and three existing algorithms are tested in the simulations. By the simulation results, our algorithm is able to perform a nice accuracy of estimation through a very low computational consumption.