摘要
以图像盲分离为背景,给出如何在小波变换下对盲源图像进行快速独立分量分析。通过探讨小波域中算法的的收敛特性,证实了小波域中算法能完美实现分离的原因,并通过实验验证了算法的有效性。
Takes the blind separation of image as background, gives a fact how to make rapid independent component analysis under wavelet transform. Through exploring the convergence of the algorithm in the Wavelet domain, confirms the reasons that FastICA can achieve perfect separation in the Wavelet domain. The experiment proves the effectiveness of the algorithm.
出处
《现代计算机》
2008年第9期73-76,共4页
Modern Computer
