Summary
International Symposium on Nonlinear Theory and Its Applications
2022
Session Number:B4L-B
Session:
Number:B4L-B-02
Hierarchical Lossless Compression Method for Color Images Using CNN Predictors
Hideharu Toda , Hisashi Aomori , Tsuyoshi Otake , Ichiro Matsuda , Susumu Itoh,
pp.374-375
Publication Date:12/12/2022
Online ISSN:2188-5079
DOI:10.34385/proc.71.B4L-B-02
PDF download (5.1MB)
Summary:
In this paper, we propose the hierarchical lossless compression method for color images using CNN predictors. In general, the selection of color space in encoding is important in terms of compression ratio. As opposed to the RGB color space, the YCoCg-R color space is known to have good coding efficiency, although it has the disadvantage of expanding the dynamic range of the color difference signals. In addition, the histogram packing technique is employed to deal with the expanded dynamic range. The experimental results on test images show that the proposed method achieves better coding performance than conventional methods.