Digital Image Restoration Using Posterior Distribution and Updating Pixel by Self Threshold

Author

Adi, Kusworo ; Mengko, Tati L R ; Suksmono, Andriyan Bayu ; Danudirdjo, Donny

Author_Institution

Sch. of Electr. Eng. & Informatics, Inst. Teknologi Bandung

fYear

2006

Firstpage

5707

Lastpage

5710

Abstract

Metropolis Hastings Markov Chain Monte Carlo (MHMCMC) algorithm has been applied to simulate digital image restoration using posterior distribution and pixels update with self threshold method. This method was implemented to Lena image degraded with 0.15 th order salt and pepper noise by the variation of temperature at 1.5, 2.5 and 4.5. The result of digital image restoration in gray scale at temperature 1.5 is good image restored with Delta_SNR 13.072 dB. In this simulation, the number of Markov chains (1000 chains) and iteration (800 iteration) are fixed parameter

Keywords

Markov processes; Monte Carlo methods; image restoration; image segmentation; statistical distributions; Lena image; Markov chains; Metropolis Hastings Markov Chain Monte Carlo algorithm; digital image restoration; fixed parameter iteration; pixels update; posterior distribution; self threshold; Degradation; Digital images; Image restoration; Low pass filters; Monte Carlo methods; Nonlinear filters; Pixel; Random variables; Stochastic resonance; Temperature; MHMCMC algorithm; Posteriori Distribution; Self Threshold;

fLanguage

English

Publisher

ieee

Conference_Titel

SICE-ICASE, 2006. International Joint Conference

Conference_Location

Busan

Print_ISBN

89-950038-4-7

Electronic_ISBN

89-950038-5-5

Type

conf

DOI

10.1109/SICE.2006.314638

Filename

4108595