• DocumentCode
    2730845
  • Title

    Applicability of DNA computing algorithm solving image recognition in intelligent visual mechanics

  • Author

    Tsuboi, Yusei ; Ono, Osiamu

  • Author_Institution
    Dept. of Electr. & Electron. Eng., Meiji Univ., Kanagawa, Japan
  • Volume
    3
  • fYear
    2003
  • fDate
    2-6 Nov. 2003
  • Abstract
    We propose an algorithm to solve image recognition problems by using stickiness of DNA molecules, and show an applicability of DNA computing in engineering research field. In normal silicon-based computer, this algorithm is not being selected because huge and parallel operations are needed. There has been a lot of processing method that several features of a model are extracted for recognizing images of it. In this paper, the features of a model for searching object construct a corresponded network. Nodes and branches in this network are encoded to DNA sequences considering complementarity of DNA, and also features of other models in search image are also encoded to them. These DNA molecules are put into hypothetical test tubes, and hybridization of DNA molecules represents where the object image exists in the several search images. DNA molecules can be used as a parallel searching computer for image recognition.
  • Keywords
    DNA; biocomputing; feature extraction; image recognition; parallel machines; pattern matching; DNA computing algorithm; DNA molecules; engineering research field; hypothetical test tubes; image recognition; intelligent visual mechanics; parallel computer; pattern matching; search images; silicon-based computer; Computational intelligence; Concurrent computing; Control systems; DNA computing; Feature extraction; Image recognition; Laboratories; Neural networks; Sequences; Testing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Industrial Electronics Society, 2003. IECON '03. The 29th Annual Conference of the IEEE
  • Print_ISBN
    0-7803-7906-3
  • Type

    conf

  • DOI
    10.1109/IECON.2003.1280691
  • Filename
    1280691