Title of article
NEAR-LOSSLESS IMAGE COMPRESSION BASED ON TWO STAGS CONSTRAINED VQ
Author/Authors
Onsi, H. M. Cairo University - Faculty of Computers and Information - Information Technology Department, Egypt , Alsayed, K. M. Cairo University - Faculty of Computers and Information - Information Technology Department, Egypt , Al Embaby, A. Cairo University - Faculty of Computers and Information - Information Technology Department, Egypt
From page
105
To page
118
Abstract
In this paper a Near-lossless image compression method based on Vector quantizer with constrained error is proposed There are two stages of Constrained Vector Quantizer (CVQ) are used, in the first stage the input image are quantized but under constrain that is no pixel error should exceed a certain error (VQ Error). The output residual from the first stage will be an input to the second stage of CVQ with VQ Error of the first stage reduced to half The reconstructed near lossless image can differ from the original one within a pixelwise error tolerance. This property is used to convert the histogram of the output residual from the second stage of CVQ; to a new histogram which is proved to have minimum entropy. Hence, new residuals matrix is formed which has minimum entropy and high spatial correlation among its pixels and can efficiently be compressed. Simulation results show the effectiveness of this compression algorithm
Keywords
Near , lossless , Local packing histogram , Constrained Error Vector Quantizer , arithmetic coding , histogram
Journal title
International Journal of Intelligent Computing and Information Sciences
Journal title
International Journal of Intelligent Computing and Information Sciences
Record number
2565484
Link To Document