Image Denoising Based on Fuzzy in Wavelet Domain

Commonly, wavelet coefficients of image are modeled as zero-mean Gaussian random variable with high local correlation. The variance estimation of wavelet coefficients is the key to get better performance for image denoising. In this paper, a fuzzy function is defined for the local variance estimation of wavelet coefficients. Based on this fuzzy function, the variance is estimated adoptively following the local feature of image. Experiments show that the proposed method has better performance in noise suppression and edges preservation, compared with other denoising methods