• DocumentCode
    1673798
  • Title

    Enhanced JPEG compression of documents

  • Author

    Prakash, Ravi ; Mitchell, Joan L. ; Stepneski, David A.

  • Author_Institution
    IBM Payment Solutions, Charlotte, NC, USA
  • Volume
    3
  • fYear
    2001
  • fDate
    6/23/1905 12:00:00 AM
  • Firstpage
    494
  • Abstract
    The JPEG standard was designed for compression of photographic digital images, but it also works well on digitized documents with only a limited number of shades of gray. For documents in which compression and legibility are more important than preserving all of the intermediate values, preprocessing the images to reduce their dynamic range can enhance JPEG compression, as it selectively discards some noise. If modified quantization tables are substituted for the encoding quantization tables in the JPEG compressed data stream, an unchanged JPEG decoder can restore the dynamic range and increase image contrast. Graphs of the compressed size in bytes vs. dynamic range after the application of different dynamic-range-reduction techniques are given for both Huffman coding and arithmetic coding. Examples of the reconstructed front and back sides of a check with normal processing and enhanced compression are shown
  • Keywords
    Huffman codes; arithmetic codes; data compression; document image processing; image coding; image reconstruction; Huffman coding; arithmetic coding; digitized documents; dynamic range reduction; encoding quantization tables; enhanced JPEG compression; Decoding; Digital images; Dynamic range; Huffman coding; Image coding; Image restoration; Noise reduction; Quantization; Streaming media; Transform coding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Processing, 2001. Proceedings. 2001 International Conference on
  • Conference_Location
    Thessaloniki
  • Print_ISBN
    0-7803-6725-1
  • Type

    conf

  • DOI
    10.1109/ICIP.2001.958159
  • Filename
    958159