A deconvolution approach to estimation of a common shape in a shifted curves model

This paper considers the problem of adaptive estimation of a mean pattern in a randomly shifted curve model. We show that this problem can be transformed into a linear inverse problem, where the density of the random shifts plays the role of a convolution operator. An adaptive estimator of the mean pattern, based on wavelet thresholding is proposed. We study its consistency for the quadratic risk as the number of observed curves tends to infinity, and this estimator is shown to achieve a near-minimax rate of convergence over a large class of Besov balls. This rate depends both on the smoothness of the common shape of the curves and on the decay of the Fourier coefficients of the density of the random shifts. Hence, this paper makes a connection between mean pattern estimation and the statistical analysis of linear inverse problems, which is a new point of view on curve registration and image warping problems. We also provide a new method to estimate the unknown random shifts between curves. Some numerical experiments are given to illustrate the performances of our approach and to compare them with another algorithm existing in the literature.

[1]  Theofanis Sapatinas,et al.  Functional deconvolution in a periodic setting: Uniform case , 2007 .

[2]  Fabrice Gamboa,et al.  Semi-parametric estimation of shifts , 2007, 0712.1936.

[3]  M. Fréchet.  Les éléments aléatoires de nature quelconque dans un espace distancié , 1948 .

[4]  D. Donoho.  Nonlinear Solution of Linear Inverse Problems by Wavelet–Vaguelette Decomposition , 1995 .

[5]  T. Gasser,et al.  Self‐modelling warping functions , 2004 .

[6]  Michael H. Neumann,et al.  On the effect of estimating the error density in nonparametric deconvolution , 1997 .

[7]  A. Tsybakov,et al.  Wavelets, approximation, and statistical applications , 1998 .

[8]  Myriam Vimond,et al.  Efficient estimation for a subclass of shape invariant models , 2010, 1010.0796.

[9]  Jean-Michel Loubes,et al.  Estimation of the distribution of random shifts deformation , 2008, 0812.3253.

[10]  J. Ramsay,et al.  Curve registration , 2018, Oxford Handbooks Online.

[11]  E. Kolaczyk.  Wavelet Methods For The Inversion Of Certain Homogeneous Linear Operators In The Presence Of Noisy D , 1994 .

[12]  I. Johnstone,et al.  Wavelet Shrinkage: Asymptopia? , 1995 .

[13]  T. Gasser,et al.  Convergence and consistency results for self-modeling nonlinear regression , 1988 .

[14]  Alain Trouvé,et al.  Bayesian template estimation in computational anatomy , 2008, NeuroImage.

[15]  K. Mardia,et al.  A penalized likelihood approach to image warping , 2001 .

[16]  Fabrice Gamboa,et al.  Estimation of Translation, Rotation, and Scaling between Noisy Images Using the Fourier--Mellin Transform , 2009, SIAM J. Imaging Sci..

[17]  R. Bhattacharya,et al.  Large sample theory of intrinsic and extrinsic sample means on manifolds--II , 2005, math/0507423.

[18]  Marc Hoffmann,et al.  Nonlinear estimation for linear inverse problems with error in the operator , 2008, 0803.1956.

[19]  Michael Stewart,et al.  The WaveD Transform in R: Performs Fast Translation-Invariant Wavelet Deconvolution , 2007 .

[20]  Vladimir Koltchinskii,et al.  On inverse problems with unknown operators , 2001, IEEE Trans. Inf. Theory.

[21]  Marc Raimondo,et al.  Wavelet Deconvolution With Noisy Eigenvalues , 2007, IEEE Transactions on Signal Processing.

[22]  Henry W. Altland,et al.  Applied Functional Data Analysis , 2003, Technometrics.

[23]  A. Tsybakov,et al.  Minimax theory of image reconstruction , 1993 .

[24]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[25]  Y. Meyer.  Wavelets and Operators , 1993 .

[26]  T. Willer.  Deconvolution in white noise with a random blurring function , 2005, math/0505142.

[27]  Yaacov Ritov,et al.  Semiparametric Curve Alignment and Shift Density Estimation for Biological Data , 2008, IEEE Transactions on Signal Processing.

[28]  H. Müller,et al.  Functional Convex Averaging and Synchronization for Time-Warped Random Curves , 2004 .

[29]  R. Gill,et al.  Applications of the van Trees inequality : a Bayesian Cramr-Rao bound , 1995 .

[30]  B. Vidakovic,et al.  Adaptive wavelet estimator for nonparametric density deconvolution , 1999 .

[31]  T. Gasser,et al.  Alignment of curves by dynamic time warping , 1997 .

[32]  Ulf Grenander,et al.  General Pattern Theory: A Mathematical Study of Regular Structures , 1993 .

[33]  H. Rosenthal.  On the Span in Lp of Sequences of Independent Random Variables (II) , 1972 .

[34]  Gerard Kerkyacharian,et al.  Wavelet deconvolution in a periodic setting , 2004 .

[35]  B. Silverman,et al.  Functional Data Analysis , 1997 .

[36]  Y. Amit,et al.  Towards a coherent statistical framework for dense deformable template estimation , 2007 .

[37]  W. W. Daniel.  Applied Nonparametric Statistics , 1979 .

[38]  T. Gasser,et al.  Searching for Structure in Curve Samples , 1995 .

[39]  Jean-Michel Loubes,et al.  Statistical M-Estimation and Consistency in Large Deformable Models for Image Warping , 2009, Journal of Mathematical Imaging and Vision.

[40]  Thomas Trigano,et al.  Semiparametric density estimation of shifts between curves , 2008 .

[41]  L. Rogers.  Stochastic differential equations and diffusion processes: Nobuyuki Ikeda and Shinzo Watanabe North-Holland, Amsterdam, 1981, xiv + 464 pages, Dfl.175.00 , 1982 .

[42]  I. Ibragimov,et al.  Norms of Gaussian sample functions , 1976 .

[43]  Jérémie Bigot.  Landmark-Based Registration of Curves via the Continuous Wavelet Transform , 2006 .

[44]  Birgitte B. Rønn,et al.  Nonparametric maximum likelihood estimation for shifted curves , 2001 .

[45]  T. Gasser,et al.  Statistical Tools to Analyze Data Representing a Sample of Curves , 1992 .

[46]  A. Tsybakov,et al.  Oracle inequalities for inverse problems , 2002 .