• DocumentCode
    2898327
  • Title

    A New Method Based on Mid -Point of Common Vertical Line to Rebuild 3D Spatial Point

  • Author

    Zhang, Mei ; Wen, Jing-hua ; Zhang, Zu-xun ; Zhang, Jian-qing

  • Author_Institution
    Sch. of Remote Sensing & Inf. Eng., Wuhan Univ.
  • fYear
    2006
  • fDate
    13-16 Aug. 2006
  • Firstpage
    3833
  • Lastpage
    3836
  • Abstract
    In the system of computer vision, we can take two cameras to synchronously capture two digital images of scenery from different angle, and then rebuild 3D shape and position of spatial scenery by computer. It is furthest basic to rebuild 3D spatial point P with method based on stereo vision. Theoretically we can compute 3D world coordinate of spatial point P by the method according intersection of two spatial lines emitted from the same point P, but in practical application, they don´t intersect because of checking error of corresponding points p1 and p2 of spatial point P in two images. This paper puts forward a new method which takes mid-point of common vertical line of two spatial lines not intersecting as the solution of spatial point P, moreover it validates this solution is classic in the meaning of making mean square error least. This new method is simple than method of least squares, furthermore it can satisfies mean square error least
  • Keywords
    computational geometry; computer vision; image reconstruction; least mean squares methods; stereo image processing; 3D spatial point rebuilding; common vertical line; computer vision; least mean square error; stereo vision; Application software; Computer vision; Cybernetics; Digital cameras; Digital images; Forward contracts; Least squares methods; Machine learning; Mean square error methods; Remote sensing; Shape; Stereo vision; 3D rebuilding; Common vertical line; Method of least square; Stereo Vision;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Machine Learning and Cybernetics, 2006 International Conference on
  • Conference_Location
    Dalian, China
  • Print_ISBN
    1-4244-0061-9
  • Type

    conf

  • DOI
    10.1109/ICMLC.2006.258693
  • Filename
    4028739