• DocumentCode
    1936144
  • Title

    A New Type of Image-Based Key

  • Author

    Kirchoff, Bruce ; Remington, David ; Lixin Fu ; Sadri, Fereidoon

  • Author_Institution
    Dept. of Biol., UNC Greensboro, Greensboro, NC
  • Volume
    1
  • fYear
    2008
  • fDate
    27-30 May 2008
  • Firstpage
    825
  • Lastpage
    829
  • Abstract
    Keys are character based tools for plant identification. They are based on the decomposition of the plant into very small, atomistic parts. These parts are described with the technical and often arcane terminology of plant taxonomy. Even the best electronic keys (Delta, Lucid) make use of this terminology. Keys are not based on pattern recognition, the forte of visual experts. Instead they demand that the user look at the plant as if it consisted of a series of isolated parts that are classified by name. Keys would be more effective if they were visually based. They would be easier to use for visual experts because accurate perception is their providence. They would also be easier to use for novices because they would not depend on knowledge of arcane terminology. This paper proposed an innovative image-based key system for species recognition.
  • Keywords
    Bayes methods; biology computing; botany; expert systems; image classification; learning (artificial intelligence); Bayesian statistical framework; expert pattern recognition; innovative image-based key system; plant decomposition; plant identification; species identification; species recognition; user interactive learning system; visual experts; Biology; Biomedical engineering; Biomedical informatics; Computer science; Image recognition; Pattern recognition; Plants (biology); Shape; Taxonomy; Terminology; holistic; image key;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    BioMedical Engineering and Informatics, 2008. BMEI 2008. International Conference on
  • Conference_Location
    Sanya
  • Print_ISBN
    978-0-7695-3118-2
  • Type

    conf

  • DOI
    10.1109/BMEI.2008.301
  • Filename
    4548786