Recovering convex boundaries from blurred and noisy observations. (English) Zbl 1113.62116

Summary: We consider the problem of estimating convex boundaries from blurred and noisy observations. In our model, the convolution of an intensity function \(f\) is observed with additive Gaussian white noise. The function \(f\) is assumed to have convex support \(G\) whose boundary is to be recovered. Rather than directly estimating the intensity function, we develop a procedure which is based on estimating the support function of the set \(G\). This approach is closely related to the method of geometric hyperplane probing, a well-known technique in computer vision applications. We establish bounds that reveal how the estimation accuracy depends on the ill-posedness of the convolution operator and the behavior of the intensity function near the boundary.


62M40 Random fields; image analysis
62G05 Nonparametric estimation
62H35 Image analysis in multivariate analysis
Full Text: DOI arXiv


[1] Anderssen, R. S. (1980). On the use of linear functionals for Abel-type integral equations in applications. In The Application and Numerical Solution of Integral Equations (R. S. Anderssen, F. R. de Hoog and M. A. Lucas, eds.) 195–221. Nijhoff, The Hague. · Zbl 0444.65084
[2] Aubin, J.-P. (1979). Applied Functional Analysis . Wiley, New York. · Zbl 0424.46001
[3] Baddeley, A. J. (1992). Errors in binary images and an \(L^p\) version of the Hausdorff metric. Nieuw Arch. Wisk. (4) 10 157–183. · Zbl 0788.68155
[4] Bertero, M. and Boccacci, P. (1998). Introduction to Inverse Problems in Imaging . Institute of Physics, Bristol and Philadelphia. · Zbl 0914.65060
[5] Brandolini, L., Rigoli, M. and Travaglini, G. (1998). Average decay of Fourier transforms and geometry of convex sets. Rev. Mat. Iberoamericana 14 519–560. · Zbl 0943.42004
[6] Bruna, J., Nagel, A. and Wainger, S. (1988). Convex hypersurfaces and Fourier transforms. Ann. of Math. ( 2 ) 127 333–365. JSTOR: · Zbl 0666.42010
[7] Candès, E. J. and Donoho, D. L. (2002). Recovering edges in ill-posed inverse problems: Optimality of curvelet frames. Ann. Statist. 30 784–842. · Zbl 1101.62335
[8] Donoho, D. L. (1999). Wedgelets: Nearly minimax estimation of edges. Ann. Statist. 27 859–897. · Zbl 0957.62029
[9] Donoho, D. L. and Low, M. G. (1992). Renormalization exponents and optimal pointwise rates of convergence. Ann. Statist. 20 944–970. · Zbl 0797.62032
[10] Efromovich, S. (1997). Robust and efficient recovery of a signal passed through a filter and then contaminated by non-Gaussian noise. IEEE Trans. Inform. Theory 43 1184–1191. · Zbl 0881.93081
[11] Ermakov, M. (1989). Minimax estimation of the solution of an ill-posed convolution type problem. Problems Inform. Transmission 25 191–200. · Zbl 0714.62006
[12] Gihman, I. I. and Skorohod, A. V. (1974). The Theory of Stochastic Processes 1 . Springer, Berlin. · Zbl 0291.60019
[13] Golberg, M. (1979). A method of adjoints for solving some ill-posed equations of the first kind. Appl. Math. Comput. 5 123–129. · Zbl 0397.65037
[14] Goldenshluger, A. (1999). On pointwise adaptive nonparametric deconvolution. Bernoulli 5 907–925. · Zbl 0953.62033
[15] Goldenshluger, A. and Spokoiny, V. (2004). On the shape-from-moments problem and recovering edges from noisy Radon data. Probab. Theory Related Fields 128 123–140. · Zbl 1035.62001
[16] Goldenshluger, A., Tsybakov, A. B. and Zeevi, A. (2006). Optimal change-point estimation from indirect observations. Ann. Statist. 34 350–372. · Zbl 1091.62021
[17] Hall, P. (1990). Optimal convergence rates in signal recovery. Ann. Probab. 18 887–900. · Zbl 0709.60048
[18] Hall, P. and Koch, I. (1990). On continuous image models and image analysis in the presence of correlated noise. Adv. in Appl. Probab. 22 332–349. JSTOR: · Zbl 0707.60038
[19] Hall, P., Nussbaum, M. and Stern, S. E. (1997). On the estimation of a support curve of indeterminate sharpness. J. Multivariate Anal. 62 204–232. · Zbl 0890.62029
[20] Hall, P. and Raimondo, M. (1998). On global performance of approximations to smooth curves using gridded data. Ann. Statist. 26 2206–2217. · Zbl 0933.62026
[21] Härdle, W., Park, B. U. and Tsybakov, A. B. (1995). Estimation of non-sharp support boundaries. J. Multivariate Anal. 55 205–218. · Zbl 0863.62030
[22] Korostelëv, A. P. and Tsybakov, A. B. (1993). Minimax Theory of Image Reconstruction . Lecture Notes in Statist. 82 . Springer, New York. · Zbl 0833.62039
[23] Lighthill, M. J. (1958). Introduction to Fourier Analysis and Generalised Functions . Cambridge Univ. Press. · Zbl 0078.11203
[24] Lindenbaum, M. and Bruckstein, A. M. (1994). Blind approximation of planar convex sets. IEEE Trans. Robotics and Automation 10 517–529.
[25] Mammen, E. and Tsybakov, A. (1995). Asymptotical minimax recovery of sets with smooth boundaries. Ann. Statist. 23 502–524. · Zbl 0834.62038
[26] Müller, H.-G. and Song, K. S. (1994). Maximin estimation of multidimensional boundaries. J. Multivariate Anal. 50 265–281. · Zbl 0798.62053
[27] Neumann, M. H. (1997). Optimal change-point estimation in inverse problems. Scand. J. Statist. 24 503–521. · Zbl 0902.62048
[28] Richards, I. and Youn, H. (1990). Theory of Distributions : A Nontechnical Introduction . Cambridge Univ. Press. · Zbl 0697.46014
[29] Schneider, R. (1993). Convex Bodies : The Brunn–Minkowski Theory . Cambridge Univ. Press. · Zbl 0798.52001
[30] Skiena, S. S. (1992). Interactive reconstruction via geometric probing. Proc. IEEE 80 1364–1383.
[31] Tsybakov, A. (1994). Multidimensional change-point problems and boundary estimation. In Change-Point Problems (E. Carlstein, H.-G. Müller and D. Siegmund, eds.) 317–329. IMS, Hayward, CA. · Zbl 1163.62331
[32] van der Vaart, A. and Wellner, J. (1996). Weak Convergence and Empirical Processes : With Applications to Statistics . Springer, New York. · Zbl 0862.60002
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.