• DocumentCode
    3286990
  • Title

    Quantifying preferential flows in porous soils: An original imaging and image processing procedure

  • Author

    Delmas, P. ; Duwig, C. ; Marquez, J. ; Prado, B.

  • Author_Institution
    Dept. of Comput. Sci., Univ. of Auckland, Auckland, New Zealand
  • fYear
    2010
  • fDate
    8-9 Nov. 2010
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    With the increasing accessibility of image acquisition techniques, image processing has become an essential component of soil science. Here, Image Processing is specifically used to the accurate prediction of solute transport through soils. To do so, dye tracers are leeched into soil cores to visualize active flow paths in cross sections of soil. We have derived a cheap alternative, using off-the-shelf equipments, as compared to earlier experiments. We still managed to compensate for most of the images acquisition process noise and provided a simple yet correctly correlated data analysis and dye concentration estimation. We tested the method using a fluorescent dye (pyranine) in intact soil cores of a Mexican allophonic (volcanic) soil. The current application of this procedure is the estimation of the pharmaceutical compounds (mostly antibiotics) contamination of agricultural soils irrigated by waste water in the northern central region surrounding Mexico City.
  • Keywords
    agriculture; contamination; data acquisition; data analysis; environmental science computing; image processing; soil; soil pollution; Mexican allophonic soil; Mexico City; agricultural soil contamination; antibiotics; data analysis; dye concentration estimation; dye tracers; fluorescent dye; image acquisition technique; image processing procedure; pharmaceutical compound; porous soil; preferential flow; soil science; solute transport prediction; wastewater; Calibration; Cameras; Image color analysis; Lighting; Optical distortion; Soil; calibration; dye tracer; principal component analysis; soil science;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image and Vision Computing New Zealand (IVCNZ), 2010 25th International Conference of
  • Conference_Location
    Queenstown
  • ISSN
    2151-2191
  • Print_ISBN
    978-1-4244-9629-7
  • Type

    conf

  • DOI
    10.1109/IVCNZ.2010.6148871
  • Filename
    6148871