Studying the influence of search rule and context shape in filtering impulse noise images with Markov chains

The paper presents an improved context-based denoising method for gray scale images affected by impulse noise. The proposed algorithm is using Markov chains to replace the detected noise with the intensity having the highest number of occurrences in similar contexts. The context of a noisy pixel consists in its neighbor pixels and is searched in a larger but limited surrounding area. We have analyzed different search methods and different context shapes. The experimental results obtained on the test images have shown that the most efficient model applies the search in form of   “*”  of contexts in form of   “+”. Besides the better denoising performance obtained on all the noise levels, the computational time has been also significantly improved with respect to our previous context-based filter which applied full search of full context. We have also compared this improved Markov filter with other denoising techniques existing in the literature, most of them being significantly outperformed.

[1]  Xinghuo Yu,et al.  Adaptive progressive filter to remove impulse noise in highly corrupted color images , 2013, Signal Image Video Process..

[2]  Hossein Nezamabadi-pour,et al.  A Fast Adaptive Salt and Pepper Noise Reduction Method in Images , 2013, Circuits Syst. Signal Process..

[3]  Ashok M. Sapkal,et al.  Poisson Noise Reducing Bilateral Filter , 2016 .

[4]  David A. Clausi,et al.  Stochastic image denoising based on Markov-chain Monte Carlo sampling , 2011, 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]  David Malah,et al.  Non-local means denoising using a content-based search region and dissimilarity kernel , 2013, 2013 8th International Symposium on Image and Signal Processing and Analysis (ISPA).

[7]  Marc-Olivier Killijian,et al.  Next place prediction using mobility Markov chains , 2012, MPM '12.

[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]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[10]  Yicong Zhou,et al.  A new weighted mean filter with a two-phase detector for removing impulse noise , 2015, Inf. Sci..

[11]  Jukka Corander,et al.  Bayesian clustering of DNA sequences using Markov chains and a stochastic partition model , 2014, Statistical applications in genetics and molecular biology.

[12]  Heikki Mannila,et al.  Random projection in dimensionality reduction: applications to image and text data , 2001, KDD '01.

[13]  Chung-waHo,et al.  CONVERGENCE OF NEWTON＇S METHOD FOR A MINIMIZATION PROBLEM IN IMPULSE NOISE REMOVAL , 2004 .

[14]  Arpad Gellert,et al.  Context-based prediction filtering of impulse noise images , 2016, IET Image Process..

[15]  Nor Ashidi Mat Isa,et al.  Noise Adaptive Fuzzy Switching Median Filter for Salt-and-Pepper Noise Reduction , 2010, IEEE Signal Processing Letters.

[16]  Rabab Kreidieh Ward,et al.  Retrieving information lost by image denoising , 2015, 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[17]  A. Ben Hamza,et al.  Removing Noise and Preserving Details with Relaxed Median Filters , 1999, Journal of Mathematical Imaging and Vision.

[18]  Xun Gong,et al.  Using Sorted Switching Median Filter to remove high-density impulse noises , 2013, J. Vis. Commun. Image Represent..

[19]  Arpad Gellert,et al.  Investigating a New Design Pattern for Efficient Implementation of Prediction Algorithms , 2013, J. Digit. Inf. Manag..

[20]  Chin-Hui Lee,et al.  An integrated approach to feature compensation combining particle filters and hidden Markov models for robust speech recognition , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[21]  Raymond H. Chan,et al.  CONVERGENCE OF NEWTON'S METHOD FOR A MINIMIZATION PROBLEM IN IMPULSE NOISE REMOVAL ∗1) , 2004 .

[22]  Zhou Wang,et al.  Progressive switching median filter for the removal of impulse noise from highly corrupted images , 1999 .

[23]  Arpad Gellert,et al.  Web prefetching through efficient prediction by partial matching , 2015, World Wide Web.

[24]  David J. Fleet,et al.  Stochastic Image Denoising , 2009, BMVC.

[25]  Bogdan Smolka,et al.  Robust local similarity filter for the reduction of mixed Gaussian and impulsive noise in color digital images , 2015, Signal Image Video Process..

[26]  Ondrej Krejcar,et al.  Non Destructive Defect Detection by Spectral Density Analysis , 2011, Sensors.

[27]  Rabab Kreidieh Ward,et al.  Synthesis and analysis prior algorithms for joint-sparse recovery , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).