Informatics and Applications
2025, Volume 19, Issue 2, pp 9-16
UNBIASED RISK ESTIMATE FOR THE FIRM SHRINKAGE METHOD OF SOLVING LINEAR INVERSE PROBLEMS
Abstract
Inverse statistical problems arise in such areas as astronomy, plasma physics, computational tomography, etc. In this case, the observed data usually contain noise and, therefore, it is necessary to apply noise suppression methods. In situations where the problem is related to the inversion of a linear homogeneous operator, noise suppression methods based on the wavelet transform and thresholding procedures have proven themselves to be effective. These methods are computationally efficient and adapt well to local features of signals. The most common types of thresholding are hard and soft thresholding. However, hard thresholding produces estimates with a large variance, while soft thresholding introduces additional bias. In an attempt to get rid of these drawbacks, various alternative types of thresholding have been proposed in recent years. This paper considers a thresholding procedure with two thresholds, which behaves like soft thresholding for small values of wavelet coefficients and like hard thresholding for large values. For this type of thresholding, an unbiased estimate of the mean-square risk is constructed and its statistical properties are analyzed. An algorithm for calculating the threshold that minimizes this estimate is also described.
[+] References (31)
- Johnstone, I. M., and B. W. Silverman. 1990. Speed of estimation in positron emission tomography and related inverse problems. Ann. Stat. 18(1):251-280.
- Donoho, D. 1995. Nonlinear solution of linear inverse problems by wavelet-vaguelette decomposition. Appl. Comput. Harmon. A. 2(2):101-126. doi: 10.1006/acha. 1995.1008.
- Abramovich, F., and B. W. Silverman. 1998. Wavelet decomposition approaches to statistical inverse problems. Biometrika 85(1):115-129. doi: 10.1093/biomet/85.1.115.
- Bruce, A. G., and H.-Y. Gao. 1997. WaveShrink with firm shrinkage. Stat. Sinica 7:855-874.
- Gao, H.-Y. 1998. Wavelet shrinkage denoising using the non-negative garrote. J. Comput. Graph. Stat. 7(4):469- 488. doi: 10.1080/10618600.1998.10474789.
- Chmelka, L., and J. Kozumplik. 2005. Wavelet- based Wiener filter for electrocardiogram signal denoising. Comput. Cardiol. 32:771-774. doi: 10.1109/CIC. 2005.1588218.
- Poornachandra, S., N. Kumaravel, T. K. Saravanan, and R. Somaskandan. 2005. WaveShrink using modified hyper-shrinkage function. 27th Annual Conference (International) of the IEEE Engineering in Medicine and Biology Society Proceedings. Piscataway, NJ: IEEE. 30-32. doi: 10.1109/IEMBS.2005.1616334.
- Lin, Y., and J. Cai. 2010. A new threshold function for signal denoising based on wavelet transform. Conference
(International) on Measuring Technology and Mechatronics Automation Proceedings. Piscataway, NJ: IEEE. 200-203. doi: 10.1109/ICMTMA.2010.347.
- Huang, H.-C., and T. C. M. Lee. 2010. Stabilized thresholding with generalized sure for image denoising. 17th Conference (International) on Image Processing Proceedings. Piscataway, NJ: IEEE. 1881-1884. doi: 10.1109/ ICIP.2010.5652353.
- Zhao, R.-M., and H.-M. Cui. 2015. Improved threshold denoising method based on wavelet transform. 7th Conference (International) on Modelling, Identification and Control Proceedings. Piscataway, NJ: IEEE. 7409352. 4 p. doi: 10.1109/ICMIC.2015.7409352.
- He, C., J. Xing, J. Li, Q. Yang, and R. Wang. 2015. Anew wavelet thresholding function based on hyperbolic tangent function. Math. Probl. Eng. 2015:528656. 10 p. doi: 10.1155/2015/528656.
- Priya, K. D., G. S. Rao, and P. S. Rao. 2016. Comparative analysis of wavelet thresholding techniques with wavelet-Wiener filter on ECG signal. Procedia Comput. Sci. 87:178-183. doi: 10.1016/j.procs.2016.05.145.
- He, H., andY. Tan. 2018. Anovel adaptive wavelet thresholding with identical correlation shrinkage function for ECG noise removal. Chinese J. Electron. 27(3):507-513. doi: 10.1049/cje.2018.02.006.
- Stein, C. 1981. Estimation of the mean of a multivariate normal distribution. Ann. Stat. 9(6):1135-1151. doi: 10.1214/aos/1176345632.
- Palionnaya, S.I., and O.V. Shestakov. 2022. Ispol'zovanie FDR-metoda mnozhestvennoy proverki gipotez pri obrashchenii lineynykh odnorodnykh operatorov [The use of the FDR method of multiple hypothesis testing when inverting linear homogeneous operators]. Informatika i ee Primeneniya - Inform. Appl. 16(2):44-51. doi: 10.14357/19922264220206. EDN: NBVVTW.
- Shestakov, O.V. 2022. Nesmeshchennaya otsenka riska porogovoy obrabotki s dvumya porogovymi znacheniyami [Unbiased thresholding risk estimate with two threshold values]. Informatika i ee Primeneniya - Inform. Appl. 16(4):14-19.doi: 10.14357/19922264220403. EDN: DZBVLC.
- Shestakov, O. V., and E. P. Stepanov. 2023. Nelineynaya regulyarizatsiya obrashcheniya lineynykh odnorodnykh operatorov s pomoshch'yu metoda blochnoy porogovoy obrabotki [Nonlinear regularization of the inversion of linear homogeneous operators using the block thresholding method]. Informatika i ee Primeneniya - Inform. Appl. 17(4):2-8. doi: 10.14357/19922264230401. EDN: PGKKYE.
- Vorontsov, M. O., and O.V. Shestakov. 2024. Asimptoticheskaya normal'nost' i sil'naya sostoyatel'nost' otsenki riska pri ispol'zovanii FDR-poroga v usloviyakh slaboy zavisimosti [Asymptotic normality and strong consistency of risk estimate when using the FDR threshold under weak dependence condition]. Informatika i ee Primeneniya - Inform. Appl. 18(3):69-79. doi: 10.14357/19922264240309. EDN: ZOQVTO.
- Shestakov, O. V. 2025. Reshenie obratnykh statisticheskikh zadach s pomoshch'yu metodov porogovoy obrabotki, dopuskayushchikh postroenie nesmeshchennoy otsenki srednekvadratichnogo riska [Solving inverse statistical problems using threshold processing methods that allow the construction of an unbiased estimate of the mean-square risk]. Informatika i ee Primeneniya - Inform. Appl. 19(1):74-81. doi: 10.14357/19922264250110. EDN: ZSIBDL.
- Mallat, S. 1999. A wavelet tour of signal processing. New York, NY: Academic Press. 857 p.
- Lee, N. 1997. Wavelet-vaguelette decompositions and homogeneous equations. West Lafayette, IN: Purdue University. PhD Thesis. 93 p.
- Johnstone, I. M. 1999. Wavelet shrinkage for correlated data and inverse problems adaptivity results. Stat. Sinica 9:51-83.
- Bruce, A. G., and Gao H.-Y. 1996. Understanding WaveShrink: Variance and bias estimation. Biometrika 83(4):727-745. doi: 10.1093/biomet/83.4.727.
- Donoho, D., and I. M. Johnstone. 1994. Ideal spatial adaptation via wavelet shrinkage. Biometrika 81(3):425- 455. doi: 10.1093/biomet/81.3.425.
- Donoho, D., and I. M. Johnstone. 1998. Minimax estimation via wavelet shrinkage. Ann. Stat. 26(3):879-921. doi: 10.1214/aos/1024691081.
- Jansen, M. 2001. Noise reduction by wavelet thresholding. Lecture notes in statistics ser. New York, NY: Springer. Vol. 161. 196 p.
- Marron, J. S., S. Adak, I. M. Johnstone, M. H. Neumann, and P. Patil. 1998. Exact risk analysis of wavelet regression. J. Comput. Graph. Stat. 7(3):278-309. doi: 10.1080/ 10618600.1998.10474777.
- Eroshenko, A.A., and O.V. Shestakov. 2014. Asymptotic normality of estimating risk upon the wavelet- vaguelette decomposition of a signal function in a model with correlated noise. Moscow University Computational Mathematics Cybernetics 38(3):110-117. doi: 10.3103/ S0278641914030042. EDN: UEUPKZ.
- Eroshenko, A. A., A. A. Kudryavtsev, and O. V. Shestakov. 2015. Limit distribution of a risk estimate using the vaguelette-wavelet decomposition of signals in a model with correlated noise. Moscow University Computational Mathematics Cybernetics 39(1):6-13. doi: 10.3103/ S0278641915010021. EDN: UFLCXN.
- Shestakov, O.V. 2017. Sil'naya sostoyatel'nost' otsenki srednekvadratichnoy pogreshnosti pri reshenii obratnykh statisticheskikh zadach [Strong consistency of the mean square risk estimate in the inverse statistical problems]. In - formatika i ee Primeneniya - Inform. Appl. 11(2):117-121. doi: 10.14357/19922264170213. EDN: YTYGBP.
- Eroshenko, A.A. 2015. Statisticheskie svoystva otsenok signalov i izobrazheniy pri porogovoy obrabotke koeffitsientov v veyvlet-razlozheniyakh [Statistical properties of signal and image estimates under thresholding of coefficients in wavelet decompositions]. Moscow: MSU. PhD Diss. 82 p.
[+] About this article
Title
UNBIASED RISK ESTIMATE FOR THE FIRM SHRINKAGE METHOD OF SOLVING LINEAR INVERSE PROBLEMS
Journal
Informatics and Applications
2025, Volume 19, Issue 2, pp 9-16
Cover Date
2025-07-10
DOI
10.14357/19922264250202
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
wavelets; threshold processing; linear homogeneous operator; unbiased risk estimate
Authors
O. V Shestakov  ,  ,
Author Affiliations
 Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V Lomonosov Moscow State University, 1-52 Leninskie Gory, GSP-1, Moscow 119991, Russian Federation
 Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
 Moscow Center for Fundamental and Applied Mathematics, M.V. Lomonosov Moscow State University, 1 Leninskie Gory, GSP-1, Moscow 119991, Russian Federation
|