• DocumentCode
    2161562
  • Title

    Chroma Characteristic Recognition Based Algorithm for Rapid Detection of Residues

  • Author

    Pan, Chunhua ; Xiang, Lian ; Zhu, Tonglin ; Lei, Hongtao

  • Author_Institution
    Coll. of Inf., South China Agric. Univ., Guangzhou, China
  • fYear
    2009
  • fDate
    17-19 Oct. 2009
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    In this paper, we study the algorithm to rapidly detect residues components using the reaction image produced by colloidal gold immunochromatography paper against the liquids to be tested. First, we transform images RGB color model into HSV color model, and then use principal color analysis to reduce three variables H, S, V into one Characteristic variable. Next we study the relationship between residues concentration and the Characteristic quantity by extracting the dominant color of image, calculating the first-moment of images, as well as using the weighted least squares method to determine the coefficients, we use Matlab simulation environment to test the characteristics quantity of a particular point in time for curve fitting, simulate the function relationship between the residues concentration and characteristics quantity. The experimental results show that the residue concentration calculated using this algorithm is consistent with the actual concentration.
  • Keywords
    chromatography; curve fitting; image colour analysis; image recognition; least squares approximations; Matlab simulation environment; chroma characteristic recognition; colloidal gold immunochromatography; curve fitting; image RGB color model transform; principal color analysis; rapid residue detection algorithm; residue concentration; weighted least squares method; Animals; Brightness; Character recognition; Curve fitting; Gold; Image color analysis; Image converters; Mathematical model; Safety; Testing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image and Signal Processing, 2009. CISP '09. 2nd International Congress on
  • Conference_Location
    Tianjin
  • Print_ISBN
    978-1-4244-4129-7
  • Electronic_ISBN
    978-1-4244-4131-0
  • Type

    conf

  • DOI
    10.1109/CISP.2009.5304326
  • Filename
    5304326