Piecewise linear tree-structured models for lossless image compression

Author

Slyz, Mako J. ; Neuhoff, David L.

Author_Institution

Dept. of Electr. Eng. & Comput. Sci., Michigan Univ., Ann Arbor, MI, USA

fYear

1996

fDate

Mar/Apr 1996

Firstpage

260

Lastpage

269

Abstract

This paper presents a probabilistic model for use in lossless image compressors. For each pixel the model provides a conditional distribution, which in the most simple case is a discretized Gaussian. The mean and variance of this Gaussian are determined by using that pixel´s neighbors to search a tree to find an autoregressive model, which is then applied to those same neighbors. Finally, arithmetic coding transmits the pixel. This paper also shows how to design this tree and find the distribution parameters associated with each leaf

Keywords

Gaussian distribution; arithmetic codes; autoregressive processes; data compression; image coding; piecewise-linear techniques; random processes; trees (mathematics); arithmetic coding; autoregressive model; conditional distribution; discretized Gaussian distribution; leaf distribution parameters; lossless image compression; piecewise linear tree-structured models; probabilistic model; tree design; Arithmetic; Image coding; Least squares methods; Parametric statistics; Piecewise linear techniques; Random processes; Random variables; Regions; Training data; USA Councils;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Compression Conference, 1996. DCC '96. Proceedings

Conference_Location

Snowbird, UT

ISSN

1068-0314

Print_ISBN

0-8186-7358-3

Type

conf

DOI

10.1109/DCC.1996.488331

Filename

488331