Abstract:
In this paper the authors demonstrate the effect of Sparse Decomposition on commonly used Independent component Analysis (ICA) algorithms which is a common tool for blind source separation problems. They also tried to achieve highest degree of accuracy in Blind Source Separation (BSS) of instantaneous linear mixtures of two-dimensional signals (images). They will show using simulated results that sparse decomposition combined with Kernel ICA algorithm produces the best separation results as compared to other ICA algorithms. Their work is motivated by the recent extensive use of ICA tool in image processing.
Page(s):
190-194
DOI:
DOI not available
Published:
Journal: Proceedings of International Conference on Information Communication Technologies, Volume: 27, Issue: 0, Year: 2008