Spatial-Correlative Picture Separation Approach Based on Blind-Signal-Processing Algorithm

In this paper, we apply the relative-gradient bound-component-analysis algorithm in blind signal processing to separate the blind signals mixed from five source pictures which are spatial-correlative. Neither the inverse matrix nor the covariance matrix is needed to be computed in this work. Even without whitening process, this methodology can perform well in a low time complexity and a low space complexity as compared with the conventional ones.