基于Laplace-Impact混合模型的DR图像去噪算法

DR Image Denoising Based on Laplace-Impact Mixture Model

提出一种双树复小波域中基于Laplace-Impact混合模型的DR图像去噪算法。该法利用局部方差建立的拉普拉斯-冲击模型的概率密度函数来逼近高频子带的系数分布。由于高频子带系数局部高度相关,所以在Laplace-Impact混合模型的框架下,采用最小均方误差估计可以很好消除DR图像噪声。实验结果表明,本文提出的算法与BLS-GSM模型和Laplace模型等经典去噪算法相比,对DR图像中的高斯噪声有更好的去噪效果。

A novel DR image denoising algorithm based on Laplace-Impact mixture model in dual-tree complex wavelet domain is proposed in this paper. It uses local variance to build probability density function of Laplace-Impact model fitted to the distribution of high-frequency subband coefficients well. Within Laplace-Impact framework, this paper describes a novel method for image denoising based on designing minimum mean squared error（MMSE） estimators, which relies on strong correlation between amplitudes of nearby coefficients. The experimental results show that the algorithm proposed in this paper outperforms several state-of-art denoising methods such as Bayes least squared Gaussian scale mixture and Laplace prior.