Abstract: A novel intra-scale and inter-scale correlation image denoising method based on Dual-tree Complex Wavelet Transform（DCWT） domain is presented to achieve the tradeoff between details retainment and noise removal.A neighborhood coefficient differential window is used to compute intra-scale correlations of complex wavelet coefficients in high frequency detail subimage，and intra-scale correlational state is identified according to the smallest error rate Bayesian decision-making rules.A HMT is fitted to the DCWT to describe the correlations between the coefficients across decomposition scales and mark inter-scale correlational state of complex wavelet coefficinents.The product result of intra-scale correlational state and inter-scale correlational state is looked as a new hidden state transition probability for HMT in DCWT.A set of iterative equations is developed using the Expectation-Maximization（EM） algorithm to estimate the model parameters and produce denoising images.Experimental results show that the proposed denoising algorithm is superior to the traditional filtering methods and possible to achieve an excellent balance between suppressing noise effectively and preserving as many image details and edges as possible.

摘要： 提出了双树复小波变换域尺度内和尺度间复系数相关性图像去噪新方法。该方法利用双树复小波变换的多方向性和平移不变性对图像进行多尺度分解，采用邻域复系数微分窗对其高频方向子图进行尺度内复系数相关性建模，并按最小错误率贝叶斯决策规则进行分类和状态标识；再把复系数尺度内状态标识与复小波域隐马尔可夫树相结合，从而实现降噪功能。实验结果表明，该方法在峰值信噪比指标上优于传统的滤波方法，能有效地抑制噪声的同时，对图像边缘具有较好的保护能力。