• DocumentCode
    3001078
  • Title

    A new method of circle’s center and radius detection in image processing

  • Author

    Mingzhu, Zhang ; Huanrong, Cao

  • Author_Institution
    Sch. of Mechtronics Eng., Henan Univ. of Sci. & Technol., Luoyang
  • fYear
    2008
  • fDate
    1-3 Sept. 2008
  • Firstpage
    2239
  • Lastpage
    2242
  • Abstract
    Circle detection and ascertaining its parameters are an important task in computer vision and pattern recognition. The method Hough transform is usually applied in this process, but this algorithm is too complex, slow and needs abundant memory especially when we deal with complicated background image with interferential information. According to the requirement of my research problem, a fast and simple method should be needed. In order to overcome this problem, this paper put forward a new algorithm to detect the circlepsilas center and radius. Select an initial point in the region of the circle image, and then track the pixels in the horizontal and vertical direction to two border pixels. According to the coordinates of the circle centerpsilas position and radius can be calculated. Experimental results show that it can detect the circle simply, stably and it can satisfy the real time demand.
  • Keywords
    Hough transforms; computational geometry; computer vision; image recognition; Hough transform; circle center detection; computer vision; image processing; pattern recognition; radius detection; Application software; Computer vision; Engine cylinders; Gaskets; Image processing; Logistics; Pattern recognition; Pixel; Robotic assembly; Robotics and automation; Circle detection; Hough transform; Image processing; fast algorithm;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on
  • Conference_Location
    Qingdao
  • Print_ISBN
    978-1-4244-2502-0
  • Electronic_ISBN
    978-1-4244-2503-7
  • Type

    conf

  • DOI
    10.1109/ICAL.2008.4636537
  • Filename
    4636537