New separation method of noisy mixed images based on DSD process

In allusion to the applicability of independent component analysis to noise-free signal separation and its limitation of poor effect when used for noisy signal, this paper gives a denoising strategy named DSD process which uses independent component analysis for the separation of noisy mixed images. In this scheme, wavelet transform is first used to denoise noisy images; then, independent component analysis is used for the separation of denoised images; finally, wavelet transform is used again to denoise separated images again; thus, clear images are obtained. The simulation experiment shows that this method can effectively improve the peak signal to noise ratio and correlation coefficient of the separation result of noisy mixed images and has a good effect.