1 October 2011 A note on mixture bivariate model
Author Affiliations +
Proceedings Volume 8285, International Conference on Graphic and Image Processing (ICGIP 2011); 828549 (2011) https://doi.org/10.1117/12.913261
Event: 2011 International Conference on Graphic and Image Processing, 2011, Cairo, Egypt
Abstract
L. Sendur and I. W. Selesnick suggest four jointly non-Gaussian bivariate models to characterize the dependency between a coefficient and its parent, and respectively derive the corresponding MAP estimators based on noisy wavelet coefficients in detail in [6]. Among the four models, the second is a mixture model and it is quite complicated to evaluate parameters, so L. Sendur and I.W. Selesnick didn't give a concrete method. In this letter, a concrete mixture bivariate model will be described by drawing inspiration from Model 2. Expectation maximization (EM) algorithm is employed to find the parameters of new model. The simulation results show that the values of PSNR have a bit improvement compared with Model 1. The results can be viewed as a supplementary of model 2 in [6].
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hanwen Cao, Wei Tian, Chengzhi Deng, "A note on mixture bivariate model", Proc. SPIE 8285, International Conference on Graphic and Image Processing (ICGIP 2011), 828549 (1 October 2011); doi: 10.1117/12.913261; https://doi.org/10.1117/12.913261
PROCEEDINGS
7 PAGES


SHARE
Back to Top