Title :
Improved Poisson intensity estimation: denoising application using Poisson data
Author :
Lu, H. ; Kim, Y. ; Anderson, John M M
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Florida, Gainesville, FL, USA
Abstract :
Recently, Timmermann and Nowak (1999) developed algorithms for estimating the means of independent Poisson random variables. The algorithms are based on a multiscale model where certain random variables are assumed to obey a beta-mixture density function. Timmermann and Nowak simplify the density estimation problem by assuming the beta parameters are known and only one mixture parameter is unknown. They use the observed data and the method of moments to estimate the unknown mixture parameter. Taking a different approach, we generate training data from the observed data and compute maximum likelihood estimates of all of the beta-mixture parameters. To assess the improved performance obtained by the proposed modification, we consider a denoising application using Poisson data.
Keywords :
image denoising; maximum likelihood estimation; method of moments; stochastic processes; Poisson data; Poisson random variables; beta-mixture density function; denoising application; improved Poisson intensity estimation; maximum likelihood estimation; method of moments; mixture parameter; multiscale model; Astronomy; Biomedical imaging; Density functional theory; Maximum likelihood estimation; Moment methods; Noise reduction; Parameter estimation; Random variables; Training data; Wavelet coefficients; Algorithms; Image Enhancement; Image Interpretation, Computer-Assisted; Likelihood Functions; Models, Statistical; Poisson Distribution; Reproducibility of Results; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Stochastic Processes;
Journal_Title :
Image Processing, IEEE Transactions on
DOI :
10.1109/TIP.2003.822606
