A method of image blind separation based on edge information
Dong Yao-hua · Information technology newsletter · 2007
When using Independent Component Analysis(ICA) method to separate the mixed image signals,if some images of the information source do not accord to the condition that images must be statistically independent each other,the ordinary ICA method may not obtain the accurate result.In this paper,combining with the features of image signal,a new kind of ICA method based on edge information is proposed in order to solve this problem.A set of experiments shows that this method can promote the precision of this kind of images and resist Gaussian noise.