摘要
利用图像子带内部Contourlet系数的空间相关性,在平移不变Contourlet去噪方法的基础上提出了一种随像素自适应调整的混合高斯模型去噪方法。每个系数建模为两个均值为零、方差不同的正态分布之和,利用局部贝叶斯阈值对Contourlet系数进行分类,通过当前系数邻域窗中两类系数的信息,得到大、小方差以及有关概率的模型参数估计。该方法较平移不变小波去噪和平移不变Contourlet去噪的Contourlet域降噪方法具有更好的降噪效果,进一步提高了PSNR值,降低了MSE值,获得了更好的图像恢复质量。
A new method for image de-noising using Gaussian mixture model based on Contourlet transform and translation invariance was presented.A pixel-adaptive Gaussian mixture model was proposed in which each coefficient was a mixture of two normal distributions with the same zero mean value and different variance.Contourlet coefficients were classified into two categories using local Bayesian threshold,and the model parameters such as large and small variances,related probabilities,could be estimated from the information of the two classified coefficients in a neighboring window.The experimental results indicate that the method can get better visual effect and PSNR value compared with the methods like wavelet and contourlet image denoising using the translation invariance.
出处
《山东农业大学学报(自然科学版)》
CSCD
北大核心
2012年第1期90-94,共5页
Journal of Shandong Agricultural University:Natural Science Edition