Informatics and Applications
2019, Volume 13, Issue 1, pp 49-54
INVERSION OF HOMOGENEOUS OPERATORS USING STABILIZED HARD THRESHOLDING WITH UNKNOWN NOISE VARIANCE
Abstract
When inverting linear homogeneous operators, it is necessary to use regularization methods, since observed data are usually noisy. For noise suppression, threshold processing of wavelet coefficients of the observed signal function is often used. Threshold processing has become a popular noise suppression tool due to its simplicity, computational efficiency, and ability to adapt to functions that have different degrees of regularity at different domains. The paper discusses the recently proposed stabilized hard thresholding method that eliminates the main drawbacks of soft and hard thresholding methods and studies statistical properties of this method. In the data model with an additive Gaussian noise with unknown variance, an unbiased estimate of the mean square risk is analyzed and it is shown that under certain conditions, this estimate is asymptotically normal and the variance of the limit distribution depends on the type of estimate of noise variance.
[+] References (12)
- Huang, H.-C., and T C. M. Lee. 2010. Stabilized thresh-olding with generalized sure for image denoising. IEEE 17th Conference (International) on Image Processing. IEEE. 1881-1884.
- Shestakov, O. V. 2018. Nonlinear regularization of inverse problems for linear homogeneous transforms by the stabi-lized hard thresholding. J. Math. Sci. 234(6):780-785.
- Kudryavtsev, A. A., and O. V. Shestakov 2011. Āsimptoti- ka otsenki riska pri veyglet-veyvlet razlozhenii nablyuda- emogo signala [The average risk assessment of the wavelet decomposition of the signal]. T-Comm - Telecommunica-tions and Their Application in Transport Industry 2:54-57.
- Kudryavtsev, A. A., and O.V. Shestakov 2011. Āsimp- toticheskoe raspredelenie otsenki riska porogovoy obra- botki veyglet-koeffitsientov signala pri neizvestnom urovne shuma [Asymptotic distribution of the risk estimate of the signal vaguelette coefficients thresholding at the unknown noise level]. T-Comm - Telecommunications and Their Application in Transport Industry 5:24-30.
- Abramovich, F, and B. W. Silverman. 1998. Wavelet de-composition approaches to statistical inverse problems. Biometrika 85(1):115-129.
- Mallat, S. 1999. A wavelet tour of signal processing. New York, NY: Academic Press. 857 p.
- Lee, N. 1997. Wavelet-vaguelette decompositions and ho-mogenous equations. West Lafayette, IN: Purdue Univer-sity. PhD Thesis. 103 p.
- Breiman, L. 1996. Heuristics of instability and stabilization in model selection. Ann. Stat. 24(6):2350-2383.
- Jansen, M. 2001. Noise reduction by wavelet thresholding. Lecture notes in statistics ser. New York, NY: Springer Verlag. Vol. 161. 196 p.
- Shestakov, O. V. 2012. O skorosti skhodimosti otsenki riska porogovoy obrabotki veyvlet-koeffitsientov k normal'no- mu zakonu pri ispol'zovanii robastnykh otsenok dispersii [On the rate of convergence to the normal law of risk estimate for wavelet coefficients thresholding when using robust variance estimates]. Informatika i ee Primeneniya - Inform. Appl. 6(2):122-128.
- Serfling, R. 1980. Approximation theorems of mathematical statistics. New York, NY: John Wiley & Sons. 371 p.
- Serfling, R., and S. Mazumder. 2009. Exponential prob-ability inequality and convergence results for the median absolute deviation and its modifications. Stat. Probabil. Lett. 79(16):1767-1773.
[+] About this article
Title
INVERSION OF HOMOGENEOUS OPERATORS USING STABILIZED HARD THRESHOLDING WITH UNKNOWN NOISE VARIANCE
Journal
Informatics and Applications
2019, Volume 13, Issue 1, pp 49-54
Cover Date
2019-04-30
DOI
10.14357/19922264190107
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
wavelets; threshold processing; unbiased risk estimate; asymptotic normality; strong consistency
Authors
O.V. Shestakov  , 
Author Affiliations
Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation
Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|