Summary
the 2014 International Symposium on Nonlinear Theory and its Applications
2014
Session Number:C3L-A
Session:
Number:C3L-A2
Scalable Lossless Image Compression Method Using CNN Predictors with Estimated Coding Bits Minimization Learning
Hisashi Aomori, Keisuke Takizawa, Tsuyoshi Otake, Masatoshi Sato, Ichiro Matsuda, Susumu Itoh, Mamoru Tanaka,
pp.569-572
Publication Date:2014/9/14
Online ISSN:2188-5079
DOI:10.34385/proc.46.C3L-A2
PDF download (399.2KB)
Summary:
This paper proposes a novel scalable lossless image coding scheme with pel-adaptive prediction using cellular neural network (CNN). The scalable image coding scheme is indispensable for modern digital archiving applications, since they are used by various mobile devices. Also, from the viewpoint of the optimal lossless coding, a pel-adaptive predictor enables high prediction performance. In this paper, edge-orientation predictors consist of space-variant CNN are used for the scalable image coding scheme having pel-adaptive prediction with no selection information. The effectiveness of proposed algorithm is validated by some computer simulations of various standard test images, and its performance is compared with that of other existing coding schemes.