Image denoising in contourlet domain based on a normal inverse Gaussian prior
Image denoising in contourlet domain based on a normal inverse Gaussian prior
复制标题
DOI：
10.1016/j.dsp.2010.01.006
复制
发表时间：
2010-09-01
影响因子：
2.9
通讯作者：
Jing, Xili
中科院分区：
文献类型：
--
作者：
Zhang, Xin;Jing, Xili
提出了一种基于正态逆高斯(NIG)概率密度函数(PDF)的子带轮廓波系数建模的图像去噪算法。这个PDF能够模拟轮廓波系数的重尾特性，并且局部参数能够模拟系数之间的级内相关性。在此框架下，我们描述了一种基于设计最大后验概率(MAP)的图像去噪新方法。此外，利用循环旋转算法对轮廓波变换平移不变性引起的边缘Gibbs现象进行了修正。实验结果表明，该方法能有效去除高斯白噪声，较好地保留图像边缘，提高峰值信噪比。(C)2010 Elsevier Inc.保留所有权利。
This paper presents a new image denoising algorithm based on the modeling of contourlet coefficients in each subband with a normal inverse Gaussian (NIG) probability density function (PDF). This PDF is able to model the heavy-tailed nature of contourlet coefficients and the local parameters model the intrascale dependency between the coefficients. Within this framework, we describe a novel method for image denoising based on designing maximum a posteriori (MAP). Furthermore, the cycle spinning algorithm is employed to modify the Gibbs phenomenon around edges caused by the lack of translation invariance of the contourlet transform. Experimental results prove that the new method can remove Gaussian white noise effectively, reserve image edges better and enhance the peak signal-to-noise ratio. (C) 2010 Elsevier Inc. All rights reserved.