• DocumentCode
    510313
  • Title

    The Segmentation of Overlapping Milk Somatic Cells Based on Improved Watershed Algorithm

  • Author

    Na, Su ; Heru, Xue

  • Author_Institution
    Coll. of Comput. & Inf. Eng., Inner Mongolia Agric. Univ., Hohhot, China
  • Volume
    3
  • fYear
    2009
  • fDate
    7-8 Nov. 2009
  • Firstpage
    563
  • Lastpage
    566
  • Abstract
    In view of the shape irregularity and the stack phenomenon in milk somatic cell pictures. An improvement watershed method is proposed in the paper, which is mainly used in resolving the over-segmentation problem in watershed algorithm. Its main idea is: distance transformation is used for binary image firstly, obtaining the central points in the local maximum regions (seed points), and merging the redundancy seed points. This makes the oversegmentation problem was basically solved. Finally the small areas are deleted after watershed segmentation. Experiments show that the improved watershed method can accurately and quickly separate the milk somatic cells and control the oversegmentation. The effect is obvious.
  • Keywords
    automatic optical inspection; biotechnology; dairy products; image segmentation; medical image processing; production engineering computing; quality control; binary image; distance transformation; image segmentation; improved watershed algorithm; local maximum regions; overlapping milk somatic cell segmentation; quality detection; redundancy seed points; shape irregularity; Agricultural engineering; Artificial intelligence; Computational intelligence; Corrosion; Dairy products; Educational institutions; Image segmentation; Merging; Shape; Transforms; cell separation; image segmentation; milk somatic cell; watershed algorithm;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Artificial Intelligence and Computational Intelligence, 2009. AICI '09. International Conference on
  • Conference_Location
    Shanghai
  • Print_ISBN
    978-1-4244-3835-8
  • Electronic_ISBN
    978-0-7695-3816-7
  • Type

    conf

  • DOI
    10.1109/AICI.2009.211
  • Filename
    5376820