A simple algorithm of blind source separation for noisy images
Xiaowei Zhang, Nuo Zhang, Jianming Lu, Hiroo Sekiya, and Takashi Yahagi
2004 RISP International Workshop on Nonlinear Circuits and Signal Processing (NCSP2004), pp.201-204, Mar., 2004. [pdf document]

<Abstract>

The Blind Source Separation (BSS) problem has become one of the main signal processing research areas, mainly due to its increasing application potential to many real world problems. A number of algorithms based on the Independent Component Analysis (ICA) theory have provided a solution to the BSS problems and have been successfully applied to a wide range of fields, covering diverse subjects such as signal processing, communications, underwater acoustics and image processing. In this study we show experimental results on using a modification of the fixed-point algorithm combined with Wavelet in image separation with Gaussian noise added.