Blind Image Separation Based on Wavelet Transformation and Sparse Component Analysis
Sha Dai · Beijing Youdian Xueyuan xuebao · 2010
Blind source separation is one of the hot spots of signal processing domain.Considering that mixed images can't be separated with sparse component analysis(SCA)model for image doesn't always satisfy sparse conditions,a method based on wavelet transformation and SCA(WT-SCA)is proposed to extract source images.WT is used to transform mixed images to frequency domain.SCA is used to estimate mixing matrix,and to reconstruct source images.The experiments manifest that WT-SCA can accurately and effectively extract sources from mixed images.The visual result and correlation coefficient analysis verify that,compared to classical FAST-ICA,separating precision of WT-SCA is higher,and separating effect is better.