提出了一种基于贝叶斯双变量模型（Bayesian Bivariate Model）和Contourlet 变换相结合的红外图像去噪算法。首先对含有加性高斯白噪声污染的红外图像进行Contourlet 变换，得到各尺度各方向上的Contourlet 系数；然后用贝叶斯双变量模型去挖掘图像Contourlet 系数的尺度间相关性；最后对处理后的系数进行Contourlet 反变换重构，得到去噪后的图像。实验结果表明，该方法有效地捕获了红外图像的轮廓信息，提高了图像的峰值信噪比，改善了图像的视觉效果。

Based on the combination of Bayesian Bivariate Model and Contourlet Transform, an algorithm for infrared image denoising is proposed. Firstly, in order to get Contourlet coefficients in all scales and directions, infrared image with additive white Gaussian noise is processed by Contourlet Transform.Then, Bayesian Bivariate Model is used to exploit the dependencies of the Contourlet coefficients across the scales.Finally, we perform inverse Contourlet Transform to the processed coefficients and get the denoised image. The experimental results demonstrate that the proposed method not only captures the contour information of infrared images more effectively, but also improves Peak Ratio of Signal to Noise and visual effects of the image remarkably.