Spatially adaptive wavelet thresholding with context modeling for image denoising

摘要: The method of wavelet thresholding for removing noise, or denoising, has been researched extensively due to its effectiveness and simplicity. Much the literature focused on developing best uniform threshold basis selection. However, not much done make values adaptive spatially changing statistics images. Such adaptivity can improve performance because it allows additional local information image (such as identification smooth edge regions) be incorporated into algorithm. This work proposes a based context modeling, common technique used in compression adapt coder characteristics. Each coefficient is modeled random variable generalized Gaussian distribution with an unknown parameter. Context modeling estimate parameter each coefficient, which then strategy. extended overcomplete expansion, yields better results than orthogonal transform. Experimental show that significantly superior quality lower MSE original assumed known.