Image despeckling using a non-parametric statistical model of wavelet coefficients

This paper investigates the image despeckling issue in the wavelet domain. A maximum a posteriori (MAP) estimation-based image despeckling approach is proposed by incorporating a non-parametric statistical model into a Bayesian inference framework. The proposed non-parametric model formulates the marginal distribution of wavelet coefficients. It differs from conventional parametric models in that the proposed model is automatically adapted to the observed image data, rather than imposing an assumption about the distribution of the data. Experiments are conducted to demonstrate the superior performance of the proposed approach.