Minimax Linewise Algorithm for Image Reconstruction

  • A. P. Korostelev
  • A. B. Tsybakov
Conference paper
Part of the Statistics and Computing book series (SCO)


We study the problem of estimating the edges in noisy images by linewise procedures. We show that the straightforward estimation method (naïve linewise procedure) does not attain the asymptotically minimax rate of accuracy as the number of observations tends to ∞. We propose the modified linewise procedure which has the asymptotically minimax rate.

Key words and phrases

image estimation edge estimation minimax rate of convergence linewise algorithm nonparametric regression change-point problem. 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Blake,A. and Zisserman,A.: (1987) Visual Reconstruction. MIT Press, Cambridge, MA.Google Scholar
  2. Bretagnolle,J. and Huber,C. (1979) Estimation des densités: risque minimax, Z. für Wahrscheinlichkeitstheorie und verw. Geb 47, 119–137.MathSciNetCrossRefzbMATHGoogle Scholar
  3. Geman,S. and Geman,D. (1984) Stochastic relaxation, Gibbs distribution, and Bayesian restoration of images. IEEE Trans. on PAMI, 6, 721–741.CrossRefzbMATHGoogle Scholar
  4. Girard,D. (1990) From template matching to optimal approximation by piecewise smooth curves. In: Curves and Surfaces in Computer Vision and Graphics, Proc. Conf. Santa Clara, California, 13–15 february, 1990, 174–182.Google Scholar
  5. Haralick,R.M. (1980) Edge and region analysis for digital image data.Comput. Graphics and Image Processing 12, 60–73.CrossRefGoogle Scholar
  6. Härdle,W. (1990) Applied Nonparametric Regression, Cambridge Univ. Pres.CrossRefzbMATHGoogle Scholar
  7. Huang,J.S. and Tseng, D.H. (1988) Statistical theory of edge detection. Comput. Graphics and Image Processing 43, 337–346.Google Scholar
  8. Ibragimov,I.A. and Khasminskii,R.Z. (1979) Statistical Estimation: Asymptotic Theory, Springer-Verlag, N.Y.Google Scholar
  9. Korostelev,A.P. (1991) Minimax reconstruction of two-dimensional images, Theory of Probability and its Applications 36, 153–159.MathSciNetCrossRefzbMATHGoogle Scholar
  10. Korostelev,A.P. and Tsybakov,A.B. (1992) Asymptotically minmax image reconstruction problems. In: “Topics in Nonparametric Estimation”, Khasminskii R.Z., ed., AMS, Providence, RI.Google Scholar
  11. Marr,D. (1982) Vision, W.H. Freeman and Co., San Francisco.Google Scholar
  12. Mumford,D. and Shah,J. (1989) Optimal approximation by piecewise smooth functions and associated variational problems. Comm. Pure Appl. Math. 42, 577–685.MathSciNetzbMATHGoogle Scholar
  13. Nagao,M. and Matsujama,T. (1979) Edge preserving smoothing. Graphics and Image Processing 9, 394–407.CrossRefGoogle Scholar
  14. Pratt,W.K. (1978) Digital Image Processing, J. Wiley, N.Y.Google Scholar
  15. Ripley,B.D. (1988) Statistical Inference for Spatial Processes. Cambridge Univ. Press.CrossRefGoogle Scholar
  16. Rosenfeld,A. and Kak,A.C. (1982) Digital Picture Processing. Academic Press, London.Google Scholar
  17. Shiau,J.J., Wahba,G. and Johnson,D.R. (1986) Partial spline models for the inclusion of tropopause and frontal boudary information in otherwise smooth two-and three-dimensional objective analysis. J. Atmospheric and Oceanic Technology 3, 714–725.CrossRefGoogle Scholar
  18. Stone,C.J. (1980) Optimal convergence for nonparametric estimators, Annals of Statistics 8, 1348–1360.MathSciNetCrossRefzbMATHGoogle Scholar
  19. Stone,C.J. (1982) Optimal global rates of convergence for nonparametric estimators, Annals of Statistics 10, 1040–1053.CrossRefzbMATHGoogle Scholar
  20. Titterington,D.M. (1985) Common structure of smoothing techniques in statistics. Internat. Statist. Review 53, 141–170.MathSciNetCrossRefzbMATHGoogle Scholar
  21. Torre,V. and Poggio,T. (1986) On edge detection. IEEE Trans. on PAMI 8, 147–163.CrossRefGoogle Scholar
  22. Tsybakov,A.B. (1989) Optimal estimation accuracy of nonsmooth images, Problems of Information Transmission 25, 180–191.MathSciNetzbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • A. P. Korostelev
    • 1
  • A. B. Tsybakov
    • 2
  1. 1.Institute for System AnalysisMoscowRussia
  2. 2.Institute de StatistiqueUniversité Catholique de LouvainLouvain-la-NeuveBelgium

Personalised recommendations