Contourlet domain image denoising using normal inverse gaussian distribution

A new contourlet-based method is introduced for reducing noise in images corrupted by additive white Gaussian noise. It is shown that a symmetric normal inverse Gaussian distribution is more suitable for modeling the contourlet coefficients than formerly-used generalized Gaussian distribution. To estimate the noise-free coefficients, a Bayesian maximum a posteriori estimator is developed utilizing the proposed distribution. In order to estimate the parameters of the distribution, a moment-based technique is used. The performance of the proposed method is studied using typical noise-free images corrupted with simulated noise and compared with that of the other state-of-the-art methods. It is shown that compared with other denoising techniques, the proposed method gives higher values of the peak signal-to-noise ratio and provides images of good visual quality.