Abstract
An important problem in image and signal analysis is denoising. Given data y j at locations x j , j = 1, ..., N, in space or time, the goal is to recover the original image or signal m j , j = 1, ..., N, from the noisy observations y j , j = 1, ..., N. Denoising is a special case of a function estimation problem: If m j = m(x j ) for some function m(x), we may model the data y j as real-valued random variables Y j satisfying the regression relation
where the additive noise ε j , j = 1, ..., N, is independent, identically distributed (i.i.d.) with mean \( \mathbb{E} \) εj = 0. The original denoising problem is solved by finding an estimate \( \hat m\left( x \right) \) of the regression function m(x) on some subset containing all the x j . More generally, we may allow the function arguments to be random variables X j ε ℝd themselves ending up with a regression model with stochastic design
where X j ,Y j are identically distributed, and \( \mathbb{E}\left\{ {\varepsilon _j |X_j } \right\} = 0 \) . In this case, the function m(x) to be estimated is the conditional expectation of Y j given X j = x.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
T. Bollerslev. Generalized autoregressive conditional heteroscedasticity. Journal of Econometrics, 31:301–327, 1986.
M. Brockmann, Th. Gasser, and E. Herrmann. Local adaptive bandwidth choice for kernel regression estimates. Journal of the American Statistical Association, 88:1302–1309, 1993.
T. Brox, M. Welk, G. Steidl, and J. Weickert. Equivalence results for TV diffusion and TV regularisation. In L. D. Griffin and M. Lillholm, editors, Scale-Space Methods in Computer Vision, volume 2695 of Lecture Notes in Computer Science, pages 86–100. Springer, Berlin-Heidelberg-New York, 2003.
Z. Cai, J. Fan, and R. Li. Efficient estimation and inference for varying coefficients models. Journal of the American Statistical Association, 95:888–902, 2000.
Z. Cai, J. Fan, and Q. Yao. Functional-coefficient regression models for nonlinear time series. Journal of the American Statistical Association, 95:941–956, 2000.
R. J. Carroll, D. Ruppert, and A.H. Welsh. Nonparametric estimation via local estimating equation. Journal of the American Statistical Association, 93:214–227, 1998.
P. Charbonnier, L. Blanc-Feraud, G. Aubert, and M. Barlaud. Two deterministic half-quadratic regularization algorithms for computed imaging. In Image Processing, 1994. Proceedings of ICIP-94, vol. 2, IEEE InternationalConference, volume 2, pages 168–172, 1994. ISBN 0-8186-6950-0
X. Chen and H. White. Nonparametric learning with feedback. Journal of Econometric Theory, 82:190–222, 1998.
C. K. Chu, I. K. Glad, F. Godtliebsen, and J. S. Marron. Edge-preserving smoothers for image processing. Journalof the American StatisticalAssociation, 93(442):526–541, 1998.
R. Dahlhaus. Fitting time series models to nonstationary processes. Annals of Statistics, 16:1–37, 1997.
R. Dahlhaus and S. Subba Rao. Statisticalinference of time varying arch processes. Annals of Statistics, 34:1074–1114, 2006.
R. Dahlhaus and S. Subba Rao. A recursive onlilne algorithm for the estimation of time varying arch processes. Bernoulli, 13:389–422, 2007.
S. Didas, J. Franke, J. Tadjuidje, and J. Weickert. Some asymptotics for localleast-squares regression with regularization. Report in Wirtschaftsmathematik 107, University of Kaiserslautern, 2007.
S. Didas, P. Mrázek, and J. Weickert. Energy-based image simplification with nonlocaldata and smoothness terms. In Algorithms for Approximation — Proceedings of the 5th InternationalConference, Chester, pages 51–60, Springer, Berlin, Heidelberg, New York, 2007.
S. Didas and J. Weickert. Integrodifferentialequations for continuous multiscale wavelet shrinkage. Inverse Problems and Imaging, 1:29–46, 2007.
H. Drees and C. Starica. A simple non-stationary modelfor stock returns, 2003. Preprint.
R. Engle. Autoregressive conditional heteroscedasticity with estimates of the variance of the united kingdom inflation. Econometrica, 50:987–1006, 1982.
J. Fan, M. Farmen, and I. Gijbels. Local maximum likelihood estimation and inference. Journal of the Royal Statistical Society, 60:591–608, 1998.
J. Fan and I. Gijbels. Local Polynomial Modelling and Its Applications. Chapman & Hall, London, 1996.
Y. Feng. Modelling different volatility components in high-frequency financial returns, 2002. Preprint.
J. Franke and S. Halim. A bootstrap test for comparing images in surface inspection. Preprint 150, dfg-spp 1114, University of Bremen, 2006.
J. Franke and S. Halim. Data-adaptive bandwidth selection for kernel estimates of two-dimensional regression functions with correlated errors. Report in Wirtschaftsmathematik, University of Kaiserslautern, 2007.
J. Franke and S. Halim. Wild bootstrap tests for signals and images, IEEE Signal Processing Magazine, 24(4):31–37, 2007.
P. Fryzlewicz, T. Sapatinas, and S. Subba Rao. Normalised least-squares estimation in time-varying arch models, 2007. To appear in: Annals of Statistics
Th. Gasser and H. G. Müller. Kernel estimation of regression functions. In Th. Gasser and M. Rosenblatt, editors, Smoothing Techniques for Curve Estimation, Springer, Berlin, Heidelberg, New York, 1979.
L. Giraitis, P. Kokoskza, and R. Leipus. Stationary arch models: Dependence structure and central limit theorem. Econometric Theory, 16:3–22, 2000.
L. Giraitis, R. Leipus, and D. Surgailis. Recent advances in arch modelling. In A. Kirman and G. Teyssiere, editors, Long memory in Economics, pages 3–39, Springer, Berlin, Heidelberg, New York, 2005.
W. Haerdle. Applied Nonparametric Regression. Cambridge University Press, Cambridge, 1990.
S. Halim. Spatially adaptive detection of local disturbances in time series and stochastic processes on the integer lattice Z2. PhD thesis, University of Kaiserlautern, 2005.
T. J. Hastie and R.J. Tibshirani. Varying-coefficient models (with discussion). Journal of the Royal Statistical Society, 55:757–796, 1993.
E. Herrmann, Th. Gasser, and A. Kneip. Choice of bandwidth for kernel regression when residuals are correlated. Biometrika, 79:783–795, 1992.
J. J. Koenderink and A. L. Van Doom. The structure of locally order-less images. International Journal of Computer Vision, 31(2/3):159–168, 1999.
H. Kushner and G. Yin. Stochastic Approximation and Recursive Algorithms and Applications. Springer, Berlin, Heidelberg, New York, 2003.
L. Ljung and T. Söderstrom. Theory and Practice of Recursive Identification. MIT Press, Cambridge, MA, 1983.
C. Loader. Local regression and likelihood. Springer, Berlin, Heidelberg, New York, 1999.
T. Mikosch and C. Starica. Is it really long memory we see in financial returns? In P. Embrechts, editor, Extremes and Integrated Risk Management, pages 149–168, Risk Books, London, 2000.
E. Moulines, P. Priouret, and F. Roueff. On recursive estimation for locally stationary time varying autoregressive processes. Annals of Statistics, 33:2610–2654, 2005.
P. Mrázek and J. Weickert. Rotationally invariant wavelet shrinkage. In Pattern Recognition, volume 2781 of Lecture Notes in Computer Science, pages 156–163. Springer, Berlin-Heidelberg, New York, 2003.
P. Mrázek, J. Weickert, and A. Bruhn. robust estimation and smoothing with spatial and tonal kernels. In R. Klette, R. Kozera, L. Noakes, and J. Weickert, editors, Geometric Properties for Incomplete Data, volume 31 of Computational Imaging and Vision, pages 335–352. Springer, Berlin-Heidelberg, New York, 2006.
P. Mrázek, J. Weickert, and G. Steidl. Correspondences between wavelet shrinkage and nonlinear diffusion. In Scale-Space Methods in Computer Vision, volume 2695 of Lecture Notes in Computer Science, pages 101–116. Springer, Berlin-Heidelberg, New York, 2003.
P. Mrázek, J. Weickert, and G. Steidl. Diffusion-inspired shrinkage functions and stability results for wavelet shrinkage. International Journal of Computer Vision, 64(2(3)):171–186, 2005.
P. Mrázek, J. Weickert, and G. Steidl. On iterations and scales of non-linear filters. In O. Drbohlav, editor, Proceedings of the Eighth Computer Vision Winter Workshop, pages 61–66, February 2003. Valtice, Czech Republic,. Czech Pattern Recognition Society.
H. Müller. Change-points in nonparametric regression analysis. Annals of Statistics, 20:737–761, 1992.
D. Mumford and J. Shah. Optimal approximation of piecewise smooth functions and associated variational problems. Communications on Pure and Applied Mathematics, 42:577–685, 1989.
D. Nelson. Conditional heteroskedasity in assit returns: A new approach. Econometrica, 59:347–370, 1990.
P. Perona and J. Malik. Scale space and edge detection using anisotropic diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12:629–639, 1990.
L. Pizarro, S. Didas, F. Bauer, and J. Weickert. Evaluating a general class of filters for image denoising. In B. K. Ersboll and K. S. Pedersen, editors, Image Analysis, volume 4522 of Lecture Notes in Computer Science, pages 601–610, Springer, Berlin, Heidelberg, New York, 2007.
G. Plonka and G. Steidl. A multiscale wavelet-inspired scheme for non-linear diffusion. International Journal of Wavelets, Multiresolution and Information Processing, 4:1–21, 2006.
J. Polzehl and V. Spokoiny. Adaptive weights smoothing with applications to image restorations. Journal of the Royal Statistical Society Ser. B, 62:335–354, 2000.
J. Polzehl and V. Spokoiny. Functional and dynamic magnetic resonance imaging using vector adaptive weights smoothing. Journal of the Royal Statistical Society Ser. C, 50:485–501, 2001.
J. Polzehl and V. Spokoiny. Image denoising: pointwise adaptive approach. Annals of Statistics, 31:30–57, 2003.
J. Polzehl and V. Spokoiny. Spatially adaptive regression estimation: Propagation-separation approach. Technical report, WIAS, Berlin, 2004. Preprint 998.
J. Polzehl and V. Spokoiny. Propagation-separation approach for local likelihood estimation. Theory and Related Fields, 135:335–336, 2006.
J. Polzehl and K. Tabelow. Adaptive smoothing of digital images: the r package adimpro. Journal of Statistical Software, 19(1), 2007.
J. Polzehl and K. Tabelow. fmri: A package for analyzing fmri data, 2007. To appear in: R News.
P. Qiu. Discontinuous regression surface fitting. Annals of Statistics, 26:2218–2245, 1998.
r. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, 2005. ISBN 3-900051-07-0 2005.
P. Robinson. Nonparametric estimation of time-varying parameters. In P. Hackl, editor, Statistical analysis and Forecasting of Economic Structural Change, pages 253–264. Springer, Berlin, Heidelberg, New York, 1989.
P. Robinson. Testing for strong serial correlation and dynamic conditional heteroskedasity in multiple regression. Journal of Econometrics, 47:67–78, 1991.
L. I. Rudin, S. Osher, and E. Fatemi. Nonlinear total variation based noise removal algorithms. Physica D, 60:259–268, 1992.
J. Simonoff. Smoothing Methods in Statistics. Springer, Berlin, Heidelberg, New York, 1996.
S. M. Smith and J. M. Brady. SUSAN — A new approach to low lewel image processing. International Journal of Computer Vision, 23(l):43–78, 1997.
V. Solo. The second order properties of a time series recursion. Annals of Statistics, 9:307–317, 1981.
V. Spokoiny. Estimation of a function with discontinuities via local polynomial fit with an adaptive window choice. Annals of Statistics, 26:1356–1378, 1998.
G. Steidl and J. Weickert. Relations between soft wavelet shrinkage and total variation denoising. Lecture Notes in Computer Science, 2449:198–205, 2002.
G. Steidl, J. Weickert, T. Brox, P. Mrázek, and M. Welk. On the equivalence of soft wavelet shrinkage, total variation diffusion, total variation regularization, and SIDEs. SIAM Journal on Numerical Analysis, 42(2):686–713, May 2004.
K. Tabelow, J. Polzehl, V. Spokoiny, and H. U. Voss. Diffusion tensor imaging: Structural adaptive smoothing. Technical report, WIAS, Berlin, 2007. Preprint 1232.
K. Tabelow, J. Polzehl, H. U. Voss, and V. Spokoiny. Analyzing fMRI experiments with structural adaptive smoothing procedures. Neuroimage, 33:55–62, 2006.
R. Tibshirani and T.J. Hastie. Local likelihood estimation. Journal of the American Statistical Association, 82:559–567, 1987.
A. Tikhonov. Solution of incorrectly formulated problems and the regularization method. In Soviet Mathematics Doklady, volume 4, pages 1035–1038, 1963.
C. Tomasi and R. Manduchi. Bilateral filtering for gray and color images. In ICCV’ 98: Proceedings of the Sixth International Conference on Computer Vision, pages 839–846, Washington, DC, USA, 1998. IEEE Computer Society.
H. U. Voss, K. Tabelow, J. Polzehl, O. Tchernichovsky, K. K. Maul, D. Salgado-Commissariat, D. Ballon, and S. A. Helekar. Functional MRI of the zebra finch brain during song stimulation suggests a lateralized response topography. PNAS, 104:10667–10672, 2007.
M.P. Wand and M.C. Jones. Kernel Smoothing. Chapman & Hall, London, 1995.
J. Weickert, G. Steidl, P. Mrazek, M. Welk, and T. Brox. Diffusion filters and wavelets: What can they learn from each other? In Handbook of Mathematical Models of Computer Vision, pages 3–16. Springer, Berlin Heidelberg New York, 2005.
M. Welk, G. Steidl, and J. Weickert. Locally analytic schemes: a link between diffusion filtering and wavelet shrinkage, 2007. To appear in: Applied and Computational Harmonic Analysis.
M. Welk, J. Weickert, and G. Steidl. A four-pixel scheme for singular differential equations. In Scale Space and PDE Methods in Computer Vision, volume 3459 of Lecture Notes in Computer Science, pages 585–597, Springer, Berlin, Heidelberg, New York, 2005.
H. White. Parametric statistical estimation using artifical neural networks. In Mathematical perspectives on neural networks, pages 719–775. L. Erlbaum Associates, Hilldale, NJ, 1996.
G. Winkler, V. Aurich, K. Hahn, and A. Martin. Noise reduction in images: Some recent edge-preserving methods. Pattern Recognition and Image Analysis, 9:749–766, 1999.
E. Zeidler, editor. Nonlinear Functional Analysis and Applications I: Fixed-Point Theorems. Springer-Verlag, New York, Inc., New York, NY, USA, 1986.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Franke, J. et al. (2008). Structural Adaptive Smoothing Procedures. In: Dahlhaus, R., Kurths, J., Maass, P., Timmer, J. (eds) Mathematical Methods in Signal Processing and Digital Image Analysis. Understanding Complex Systems. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-75632-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-540-75632-3_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-75631-6
Online ISBN: 978-3-540-75632-3
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)