A Quadsection Algorithm for Grammar-Based Image Compression

  • Morihiro Hayashida
    Bioinformatics Center, Institute for Chemical Research, Kyoto University
  • Peiying Ruan
    Bioinformatics Center, Institute for Chemical Research, Kyoto University
  • Tatsuya Akutsu
    Bioinformatics Center, Institute for Chemical Research, Kyoto University

この論文をさがす

抄録

For the purpose of text compression, grammar-based compression, which is to find a small grammar that generates a given data, has been well-studied. In this technical report, we apply this methodology to compression of rectangular image data. We first define a context-free rectangular image grammar (CFRIG) by extending the context-free grammar. Then we propose a quadsection type algorithm by extending a bisection type algorithm for grammar-based compression of text data. We show that our proposed algorithm approximates in polynomial time the smallest CFRIG within a factor of O(n4/3), where an input image data is of size O(n) × O(n). We also present results on computational experiments on the proposed algorithm.For the purpose of text compression, grammar-based compression, which is to find a small grammar that generates a given data, has been well-studied. In this technical report, we apply this methodology to compression of rectangular image data. We first define a context-free rectangular image grammar (CFRIG) by extending the context-free grammar. Then we propose a quadsection type algorithm by extending a bisection type algorithm for grammar-based compression of text data. We show that our proposed algorithm approximates in polynomial time the smallest CFRIG within a factor of O(n4/3), where an input image data is of size O(n) × O(n). We also present results on computational experiments on the proposed algorithm.

収録刊行物

参考文献 (9)*注記

もっと見る

詳細情報 詳細情報について

  • CRID
    1570572702598115328
  • NII論文ID
    110007993903
  • NII書誌ID
    AN10505667
  • ISSN
    09196072
  • 本文言語コード
    en
  • データソース種別
    • CiNii Articles

問題の指摘

ページトップへ