2-D Impulse Noise Suppression by Recursive Gaussian Maximum Likelihood Estimation

An effective approach termed Recursive Gaussian Maximum Likelihood Estimation (RGMLE) is developed in this paper to suppress 2-D impulse noise. And two algorithms termed RGMLE-C and RGMLE-CS are derived by using spatially-adaptive variances, which are respectively estimated based on certainty and joint certainty & similarity information. To give reliable implementation of RGMLE-C and RGMLE-CS algorithms, a novel recursion stopping strategy is proposed by evaluating the estimation error of uncorrupted pixels. Numerical experiments on different noise densities show that the proposed two algorithms can lead to significantly better results than some typical median type filters. Efficient implementation is also realized via GPU (Graphic Processing Unit)-based parallelization techniques.

[1]  B I Justusson,et al.  Median Filtering: Statistical Properties , 1981 .

[2]  M. West.  On scale mixtures of normal distributions , 1987 .

[3]  Rodrigo L. Carceroni.  Journal of Mathematical Imaging and Vision: SIBGRAPI 2006 Special Issue , 2007, Journal of Mathematical Imaging and Vision.

[4]  Gonzalo R. Arce,et al.  A general weighted median filter structure admitting negative weights , 1998, IEEE Trans. Signal Process..

[5]  David Ebenezer,et al.  A New Fast and Efficient Decision-Based Algorithm for Removal of High-Density Impulse Noises , 2007, IEEE Signal Processing Letters.

[6]  Dimitri Van De Ville,et al.  Nonlocal Means With Dimensionality Reduction and SURE-Based Parameter Selection , 2011, IEEE Transactions on Image Processing.

[7]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[8]  Raymond H. Chan,et al.  Salt-and-pepper noise removal by median-type noise detectors and detail-preserving regularization , 2005, IEEE Transactions on Image Processing.

[9]  Veerakumar Thangaraj,et al.  Removal of High Density Salt and Pepper Noise Through Modified Decision Based Unsymmetric Trimmed Median Filter , 2011, IEEE Signal Processing Letters.

[10]  Peyman Milanfar,et al.  Kernel Regression for Image Processing and Reconstruction , 2007, IEEE Transactions on Image Processing.

[11]  Michael Elad,et al.  Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries , 2006, IEEE Transactions on Image Processing.

[12]  Wufan Chen,et al.  Joint-MAP Tomographic Reconstruction with Patch Similarity Based Mixture Prior Model , 2011, Multiscale Model. Simul..

[13]  Florence Tupin,et al.  Iterative Weighted Maximum Likelihood Denoising With Probabilistic Patch-Based Weights , 2009, IEEE Transactions on Image Processing.

[14]  Jian-Feng Cai,et al.  Two-phase approach for deblurring images corrupted by impulse plus gaussian noise , 2008 .

[15]  Pradeep Kumar,et al.  Salt-and-pepper noise removal by adaptive median-based lifting filter using second-generation wavelets , 2013, Signal Image Video Process..

[16]  N. Saitou,et al.  Maximum likelihood methods. , 1990, Methods in enzymology.

[17]  Raymond H. Chan,et al.  A Detection Statistic for Random-Valued Impulse Noise , 2007, IEEE Transactions on Image Processing.

[18]  Martin J. Wainwright,et al.  Scale Mixtures of Gaussians and the Statistics of Natural Images , 1999, NIPS.

[19]  Shuqun Zhang,et al.  A new impulse detector for switching median filters , 2002, IEEE Signal Processing Letters.

[20]  Youlun Xiong,et al.  Impulse Noise Removal Using Directional Difference Based Noise Detector and Adaptive Weighted Mean Filter , 2009, IEEE Signal Processing Letters.

[21]  Christine Toumoulin,et al.  Thoracic low-dose CT image processing using an artifact suppressed large-scale nonlocal means , 2012, Physics in medicine and biology.

[22]  John Harris,et al.  Handbook of mathematics and computational science , 1998 .

[23]  Jianhua Ma,et al.  Nonlocal Prior Bayesian Tomographic Reconstruction , 2008, Journal of Mathematical Imaging and Vision.

[24]  Stan Uryasev,et al.  Maximum Likelihood Method , 2014 .

[25]  Yong-Hwan Lee,et al.  A simple formula for the variance of the median filter output with Laplacian input , 1991, IEEE Trans. Signal Process..

[26]  Richard A. Haddad,et al.  Adaptive median filters: new algorithms and results , 1995, IEEE Trans. Image Process..

[27]  Jerry D. Gibson,et al.  Handbook of Image and Video Processing , 2000 .

[28]  Chih-Hsing Lin,et al.  Switching Bilateral Filter With a Texture/Noise Detector for Universal Noise Removal , 2010, IEEE Transactions on Image Processing.

[29]  Jean-Michel Morel,et al.  A non-local algorithm for image denoising , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[30]  Amit Singer,et al.  Non-Local Euclidean Medians , 2012, IEEE Signal Processing Letters.

[31]  Kai-Kuang Ma,et al.  A switching median filter with boundary discriminative noise detection for extremely corrupted images , 2006, IEEE Trans. Image Process..

[32]  J. Astola,et al.  Fundamentals of Nonlinear Digital Filtering , 1997 .