Adaptive fractal-wavelet image denoising base on multivariate statistical model

In order to overcome the defects that fractal wavelet denoising method can't effectively protect the edges and details of image,a adaptive fractal-wavelet image denoising scheme based on multivariate statistical model is presented in this article.A multivariate statistical model whose parameters can be improved flexibly is established to accurately estimate all relevant information in this method and the parameters of model are adaptively adjusted by minimizing the residual.In order to remove noise and protect the edges and details of image,near-optimal parent subtree is found in the best subtree domain by using collage distance under moderate noise and quadtree segmentation is used to adaptive predict fractal-wavelet coding of noiseless image from the noise image.Experimental results show that this method can effectively maintain the edge features of image and retain the fine structure of image primely while noise be removed.As the prediction of wavelet fractal coding is adopted in this scheme,the algorithm structure has been optimized,and the processing speed for algorithm is faster.Therefore,it can achieve the processing speed requirements for real-time image denoising.