Proceedings Article | 28 January 2009
KEYWORDS: Image compression, RGB color model, Wavelets, Computer programming, Image quality standards, Discrete wavelet transforms, Bismuth, Image processing, JPEG2000, Medical imaging
Eliminating correlated redundancy of space and energy by using a DWT lifting scheme and reducing the complexity of
the image by using an algebraic transform among the RGB components. An improved Rice Coding algorithm, in which
presents an enumerating DWT lifting scheme that fits any size images by image renormalization has been proposed in
this paper. This algorithm has a coding and decoding process without backtracking for dealing with the pixels of an
image. It support LOCO-I and it can also be applied to Coder / Decoder. Simulation analysis indicates that the proposed
method can achieve a high image compression. Compare with Lossless-JPG, PNG(Microsoft), PNG(Rene),
PNG(Photoshop), PNG(Anix PicViewer), PNG(ACDSee), PNG(Ulead photo Explorer), JPEG2000, PNG(KoDa Inc),
SPIHT and JPEG-LS, the lossless image compression ratio improved 45%, 29%, 25%, 21%, 19%, 17%, 16%, 15%,
11%, 10.5%, 10% separately with 24 pieces of RGB image provided by KoDa Inc. Accessing the main memory in
Pentium IV,CPU2.20GHZ and 256MRAM, the coding speed of the proposed coder can be increased about 21 times than
the SPIHT and the efficiency of the performance can be increased 166% or so, the decoder's coding speed can be
increased about 17 times than the SPIHT and the efficiency of the performance can be increased 128% or so.
