• DocumentCode
    2157148
  • Title

    Research of Double-Threshold Segmentation of Brazing-Area Defect of Saw Based on Otsu and HSV Color Space

  • Author

    Xiong, Lin Lin ; Wang, Zhong ; Wang, Xin Zhong

  • Author_Institution
    State Key Lab. of Precision Meas. Technol. & Instrum., Tianjin Univ., Tianjin, China
  • fYear
    2009
  • fDate
    17-19 Oct. 2009
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    Saws are important tools of cutting process in industrial production. However there are often wasters with surface defects due to processing technique, therefore the detection of surface defects of saws is crucial to the quality. This paper concerns the segmentation of brazing-area of saw. Considering real-time request of the detection, Otsu method which is simple and rapid was used. HSV color space which is similar to visual model was chosen. Gray space Otsu algorithm and color space double-threshold segmentation based on Otsu (Wang Xiangke and Zheng Zhiqiang, 2006) were compared with method combined H-S information in HSV color space with Otsu in this paper. The results show that the algorithm in this paper has a good effect and can meet real-time demand.
  • Keywords
    brazing; cutting tools; fracture; image colour analysis; image segmentation; mechanical engineering computing; sawing; H-S information; HSV color space; Otsu method; brazing-area defect; cutting tool; double-threshold segmentation; gray space Otsu algorithm; industrial production; saw; surface defect detection; visual model; Extraterrestrial measurements; Eyes; Image analysis; Image color analysis; Image processing; Image segmentation; Laboratories; Pattern recognition; Production; Space technology;
  • 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.2009.5304160
  • Filename
    5304160