Compression of color-mapped images

A. C. Hadenfeldt, K. Sayood

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In a standard image coding scenario, pixel-to-pixel correlation nearly always exists in the data, especially if the image is a natural scene. This correlation is what allows predictive coding schemes (e.g., DPCM) to perform efficient compression. In a color-mapped image, the values stored in the pixel array are no longer directly related to the pixel intensity. Two color indices which are numerically adjacent (close) may point to two very different colors. The correlation still exists, but only via the colormap. This fact can be exploited by sorting the colour map to reintroduce the structure. In this paper we study the sorting of colormaps and show how the resulting structure can be used in both lossless and lossy compression of images.

Original languageEnglish (US)
Title of host publicationProceedings of the 1993 IEEE International Conference on Communications
PublisherPubl by IEEE
Pages537-541
Number of pages5
ISBN (Print)0780309510
StatePublished - 1993
EventProceedings of the IEEE International Conference on Communications '93 - Geneva, Switz
Duration: May 23 1993May 26 1993

Publication series

NameProceedings of the 1993 IEEE International Conference on Communications

Other

OtherProceedings of the IEEE International Conference on Communications '93
CityGeneva, Switz
Period5/23/935/26/93

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'Compression of color-mapped images'. Together they form a unique fingerprint.

Cite this