Summary
2007 International Symposium on Nonlinear Theory and its Applications
2007
Session Number:19AM1-C
Session:
Number:19AM1-C-2
Nonlinear prediction on image signals using radial basis function network
Keisuke Narisawa, Takashi Hoshino, Naoki Yabuta, Tohru Ikeguchi,
pp.381-384
Publication Date:2007/9/16
Online ISSN:2188-5079
DOI:10.34385/proc.41.19AM1-C-2
PDF download (414.2KB)
Summary:
Differential pulse code modulation (DPCM) is one of the most popular methods to compress image signals. Although the DPCM generally works well, it could be flawed, because the DPCM predicts pixel values only by a linear function. Thus, the prediction errors might become large. In this paper, we proposed a nonlinear prediction method that incorporates the DPCM and a radial basis function (RBF) network. The proposed nonlinear prediction method reduces prediction errors of the DPCM by the RBF network. We con?rmed that proposed method reduces prediction errors of the DPCM and average bit rates by numerical simulation to several standard images.