Image denoising plays an important role in digital image processing, and the number of its method is much larger than before as well. Its essence is to use the characteristics of noise and various kinds of prior information as much as possible to remove interference, keep the information of the image and improve image quality. In consideration of both the transitivity of the scale of the wavelet coefficients and the scale space aggregation, this paper proposes a new denoising algorithm based on neighborhood model and bivariate model to deal with the disadvantages of traditional algorithm.By researching on the statistical properties of wavelet coefficients of the uniform discrete curvelet transform domain, the article proposes a new bivariate model denoising algorithm to make up for the shortage of losing sight of spatial clustering in traditional bivariate model. Firstly, Considering to the correlation of wavelet coefficient of the scale, the neighborhood model is brought in to estimate the noise variance using the monte carlo method for image denoising. Secondly, in order to obtain the initial image, on the basis of the correlation of wavelet coefficient and the parent coefficient,we introduce a bivariate model of scale transitivity. Finally, initial image noise and the original noising image are used as prior information to deduce the improved bivariate model for dealing with the original noising image. And the final denoising image is acquired under the convergence condition of symmetric Kullback-Leibler divergence and the maximum number of iterations.The experimental results show that the algorithm effectively improves the quality of denoising image.This article introduces the mean square error (MSE) and peak signal-to-noise ratio (PSNR) to measure the image denoising effect. Compared with the previous denoising algorithm,the algorithm mentioned in this article retains the edge features and the image for more details to eliminate the possibility of a smoothed image. The experimental results prove the effectiveness of this method. |