×

Wavelet deconvolution in a periodic setting. (English) Zbl 1046.62039

Summary: Deconvolution problems are naturally represented in the Fourier domain, whereas thresholding in wavelet bases is known to have broad adaptivity properties. We study a method which combines both fast Fourier and fast wavelet transforms and can recover a blurred function observed in white noise with \(O\{n\log(n)^2\}\) steps. In the periodic setting, the method applies to most deconvolution problems, including certain ’boxcar’ kernels, which are important as a model of motion blur, but having poor Fourier characteristics. Asymptotic theory informs the choice of tuning parameters and yields adaptivity properties for the method over a wide class of measures of error and classes of functions.
The method is tested on simulated light detection and ranging data suggested by underwater remote sensing. Both visual and numerical results show an improvement over competing approaches. Finally, the theory behind our estimation paradigm gives a complete characterization of the ‘maxiset’ of the method: the set of functions where the method attains a near optimal rate of convergence for a variety of \(L^p\) loss functions.

MSC:

62G08 Nonparametric regression and quantile regression
65T60 Numerical methods for wavelets
65T50 Numerical methods for discrete and fast Fourier transforms
65C60 Computational problems in statistics (MSC2010)

Software:

WaveLab
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1093/biomet/85.1.115 · Zbl 0908.62095
[2] Bertero M., Introduction to Inverse Problems in Imaging (1998) · Zbl 0914.65060
[3] J. Buckheit, S. Chen, D. Donoho, and I. Johnstone (1995 ) Wavelab reference manual . Stanford University, Stanford. (Available fromhttp://www-stat.stanford.edu/wavelab/.)
[4] Coifman R., Lect. Notes Statist 103 pp 125– (1995)
[5] DOI: 10.1006/acha.1995.1008 · Zbl 0826.65117
[6] Donoho D. L., J. R. Statist. Soc. 57 pp 301– (1995)
[7] D. L. Donoho, and M. Raimondo (2004 ) Translation invariant deconvolution in a periodic setting . to be published. · Zbl 1071.62088
[8] Efromovich S., J. Am. Statist. Ass. 94 pp 189– (1999)
[9] DOI: 10.1109/18.986021 · Zbl 1071.94511
[10] Fefferman C., Am. J. Math. 93 pp 107– (1971)
[11] S. Harsdorf, and R. Reuter (2000 ) Stable deconvolution of noisy lidar signals . Oldenburg University, Oldenburg. (Available fromhttp://las.physik.uni-oldenburg.de/projekte/earsel/4th_workshop_paper/harsdorf.pdf.)
[12] Hernandez E., A First Course on Wavelets (1996)
[13] Jain A., Fundamentals of Digital Image Processing (1989) · Zbl 0744.68134
[14] Je Park Y., Appl. Opt. 36 pp 5158– (1997)
[15] Johnstone I. M., Ann. Statist 32 (2004)
[16] DOI: 10.1214/aos/1046294458 · Zbl 1102.62318
[17] Kerkyacharian G., Test 9 pp 283– (2000)
[18] Khinchin A. Y., Continued Fractions (1997) · JFM 63.0924.02
[19] E. Kolaczyk (1994 ) PhD Dissertation . PhD Dissertation . Department of Statistics, Stanford University, Stanford.
[20] Lang S., Introduction to Diophantine Approximations (1966) · Zbl 0144.04005
[21] DOI: 10.1109/83.892445 · Zbl 1039.68788
[22] Mallat S., A Wavelet Tour of Signal Processing (1999) · Zbl 0945.68537
[23] Meyer Y., Wavelets and Operators (1992)
[24] Meyer Y., Wavelets: Calderon-Zygmund and Multilinear Operators (1997)
[25] DOI: 10.1109/TSP.2003.821103 · Zbl 1369.94238
[26] DOI: 10.1214/aos/1017939249 · Zbl 0962.62030
[27] DOI: 10.1081/NFA-120003678 · Zbl 1006.65145
[28] Walter G. G., Wavelet and Other Orthogonal Systems with Applications (1994)
[29] Walter G., J. Integr. Eqns Appl. 11 pp 515– (1999)
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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.