• DocumentCode
    1283122
  • Title

    Ill-posed problems in early vision

  • Author

    Bertero, Mario ; Poggio, T.A. ; Torre, Vincent

  • Author_Institution
    Dept. of Phys., Istituto Nazionale di Fisica Nucl., Genova, Italy
  • Volume
    76
  • Issue
    8
  • fYear
    1988
  • fDate
    8/1/1988 12:00:00 AM
  • Firstpage
    869
  • Lastpage
    889
  • Abstract
    Mathematical results on ill-posed and ill-conditioned problems are reviewed and the formal aspects of regularization theory in the linear case are introduced. Specific topics in early vision and their regularization are then analyzed rigorously, characterizing existence, uniqueness, and stability of solutions. A fundamental difficulty that arises in almost every vision problem is scale, that is, the resolution at which to operate. Methods that have been proposed to deal with the problem include scale-space techniques that consider the behavior of the result across a continuum of scales. From the point of view of regulation theory, the concept of scale is related quite directly to the regularization parameter λ. It suggested that methods used to obtained the optimal value of λ may provide, either directly or after suitable modification, the optimal scale associated with the specific instance of certain problems
  • Keywords
    computer vision; analyzed rigorously; computer vision; concept of scale; early vision; existence; first state of processing; ill posed conditions; ill-conditioned problems; optimal scale; regularization parameter; regularization theory; regulation theory; resolution; scale-space techniques; solution existence; solution uniqueness; stability of solutions; uniqueness; vision problem; Artificial intelligence; Biomedical optical imaging; Computer vision; Image edge detection; Image motion analysis; Interpolation; Laboratories; Machine vision; Nonlinear optics; Shape; Stability analysis;
  • fLanguage
    English
  • Journal_Title
    Proceedings of the IEEE
  • Publisher
    ieee
  • ISSN
    0018-9219
  • Type

    jour

  • DOI
    10.1109/5.5962
  • Filename
    5962