独立分量分析在图像去噪方面的应用

Independent Component Analysis of Application in Image Denoising

独立分量分析(independent component analysis,ICA)是基于信号高阶统计量的盲源分离方法。在分析独立分量分析的基本模型及方法的基础上,讨论了有噪信号的独立分量分析(Noisy ICA),结合传统有噪图像分离方法与结合改进FastlCA算法有噪图像分离仿真研究进行对比。结果表明,该算法即使在高水平噪声图像中,也能够分理出比较清晰的图像。

Independent component analysis（ICA） is a signal higher-order statistics based blind source separation methods.Independent component analysis in analysis of the basic model and methods based on the discussion of a noise signal independent component analysis（Noisy ICA）,combined with traditional methods and noisy image separation algorithm combined with improved FastlCA noisy image separation simulation studies are compared.The results show that the algorithm even in high-noise images,but also be able to sort out a clearer image points.