Lossy compression of palettized images

  • Authors:
  • Yung Chen;Heidi Peterson;Walter Bender

  • Affiliations:
  • Merck & Co., Rahway, NJ;lBM T. J. Watson Research Center, Yorktown Heights, NY;MIT Media Lab, Cambridge, MA

  • Venue:
  • ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: image and multidimensional signal processing - Volume V
  • Year:
  • 1993

Quantified Score

Hi-index 0.00

Visualization

Abstract

Many digital display systems economize by rendering color images using a limited palette. Palettized images differ from continuous-tone images in two important ways: pixel values are lookup table indices instead of physical intensity values, and pixel values may be dithered for better color rendition. These characteristics reduce spatial continuity in the image, leading to high bit rates and low image quality when compressing these images using a conventional lossy coder. Furthermore, decoded pixels may take on values outside the palette, necessitating palettization of the decoded image for display. We present an algorithm for lossy compression of palettized images which uses standard compression techniques (JPEG, JBIG), and produces a palettized image at the decoder. The technique is applicable to both custom and universal palettized images, and achieves good decompressed image quality at bit rates substantially lower than lossless compression of the palettized image.