Abstract
PSF (point spread function) estimation plays an important role in blind image deconvolution. It has been shown in our previous work that minimization of the Stein’s unbiased risk estimate (SURE) – unbiased estimate of mean squared error (MSE) – could yield an accurate PSF estimate.
In this paper, we show that the PSF estimation error is upper bounded by the deconvolution accuracy and the mismatch between the assumed PSF parametric form and the underlying true one. For this reason, we incorporate the
Funding source: National Natural Science Foundation of China
Award Identifier / Grant number: 61401013
Funding statement: This work was supported by the National Natural Science Foundation of China under Grant No. 61401013.
Acknowledgements
The authors would like to thank the anonymous reviewers for their helpful comments, which substantially improved the theoretical quality of this paper.
References
[1] A. Ahmed, B. Recht and J. Romberg, Blind deconvolution using convex programming, IEEE Trans. Inform. Theory 60 (2014), no. 3, 1711–1732. 10.1109/TIT.2013.2294644Search in Google Scholar
[2] F. Alvarez and H. Attouch, An inertial proximal method for maximal monotone operators via discretization of a nonlinear oscillator with damping, Set-Valued Anal. 9 (2001), 3–11. 10.1023/A:1011253113155Search in Google Scholar
[3] H. H. Bauschke and P. L. Combettes, Convex Analysis and Monotone Operator Theory in Hilbert Spaces, CMS Books Math./OuvragesMath. SMC, Springer, New York, 2011. 10.1007/978-1-4419-9467-7Search in Google Scholar
[4] A. Beck and M. Teboulle, A fast iterative shrinkage-thresholding algorithm for linear inverse problems, SIAM J. Imaging Sci. 2 (2009), no. 1, 183–202. 10.1137/080716542Search in Google Scholar
[5] T. Blu and F. Luisier, The SURE-LET approach to image denoising, IEEE Trans. Image Process. 16 (2007), no. 11, 2778–2786. 10.1109/TIP.2007.906002Search in Google Scholar
[6] R. I. Boţ, E. R. Csetnek and C. Hendrich, Inertial Douglas-Rachford splitting for monotone inclusion problems, Appl. Math. Comput. 256 (2015), 472–487. 10.1016/j.amc.2015.01.017Search in Google Scholar
[7] A. S. Carasso, The APEX method in image sharpening and the use of low exponent Lévy stable laws, SIAM J. Appl. Math. 63 (2002), no. 2, 593–618. 10.6028/NIST.IR.6749Search in Google Scholar
[8] F. Chen and J. Ma, An empirical identification method of Gaussian blur parameter for image deblurring, IEEE Trans. Signal Process. 57 (2009), no. 7, 2467–2478. 10.1109/TSP.2009.2018358Search in Google Scholar
[9] P. L. Combettes and V. R. Wajs, Signal recovery by proximal forward-backward splitting, Multiscale Model. Simul. 4 (2005), no. 4, 1168–1200. 10.1137/050626090Search in Google Scholar
[10] A. Cornelio, E. Loli Piccolomini and J. G. Nagy, Constrained numerical optimization methods for blind deconvolution, Numer. Algorithms 65 (2014), no. 1, 23–42. 10.1007/s11075-013-9693-zSearch in Google Scholar
[11] A. Cornelio, F. Porta and M. Prato, A convergent least-squares regularized blind deconvolution approach, Appl. Math. Comput. 259 (2015), 173–186. 10.1016/j.amc.2015.02.048Search in Google Scholar
[12] A. Danelakis, M. Mitrouli and D. Triantafyllou, Blind image deconvolution using a banded matrix method, Numer. Algorithms 64 (2013), no. 1, 43–72. 10.1007/s11075-012-9654-ySearch in Google Scholar
[13] C.-A. Deledalle, S. Vaiter, J. Fadili and G. Peyré, Stein unbiased gradient estimator of the risk (SUGAR) for multiple parameter selection, SIAM J. Imaging Sci. 7 (2014), no. 4, 2448–2487. 10.1137/140968045Search in Google Scholar
[14] Y. C. Eldar, Generalized SURE for exponential families: Applications to regularization, IEEE Trans. Signal Process. 57 (2009), no. 2, 471–481. 10.1109/TSP.2008.2008212Search in Google Scholar
[15] H. Fang, Y. Shi, D. Pan and G. Zhou, Iteratively reweighted blind deconvolution for passive millimeter-wave images, Signal Process. 138 (2017), 182–194. 10.1016/j.sigpro.2017.01.021Search in Google Scholar
[16] E. Faramarzi, D. Rajan and M. P. Christensen, Unified blind method for multi-image super-resolution and single/multi-image blur deconvolution, IEEE Trans. Image Process. 22 (2013), no. 6, 2101–2114. 10.1109/TIP.2013.2237915Search in Google Scholar PubMed
[17] R. Fergus, B. Singh, A. Hertzmann, S. T. Roweis and W. T. Freeman, Removing camera shake from a single photograph, ACM Trans. Graphics 25 (2006), 787–794. 10.1145/1179352.1141956Search in Google Scholar
[18] T. Hastie, R. Tibshirani and J. Friedman, The Elements of Statistical Learning, 2nd ed., Springer, New York, 2009. 10.1007/978-0-387-84858-7Search in Google Scholar
[19] T. A. Hearn and L. Reichel, Extensions of the Justen–Ramlau blind deconvolution method, Adv. Comput. Math. 39 (2013), no. 3–4, 465–491. 10.1007/s10444-012-9290-zSearch in Google Scholar
[20] A. Levin, Y. Weiss and F. Durand, Understanding blind deconvolution algorithms, IEEE Trans. Pattern Anal. Machine Intelligence 33 (2011), no. 12, 2354–2367. 10.1109/TPAMI.2011.148Search in Google Scholar PubMed
[21] D. Li and S. Simske, Atmospheric turbulence degraded-image restoration by kurtosis minimization, IEEE Geosci. Remote Sensing Letters 6 (2009), no. 2, 244–247. 10.1109/LGRS.2008.2011569Search in Google Scholar
[22] J. Li, F. Xue and T. Blu, Fast and accurate 3D PSF computation for fluorescence microscopy, J. Optical Soc. Amer. A 34 (2017), no. 6, 1029–1034. 10.1364/JOSAA.34.001029Search in Google Scholar PubMed
[23] J. Li, F. Xue and T. Blu, Gaussian blur estimation for photon-limited images, 2017 IEEE International Conference on Image Processing (ICIP), IEEE Press, Piscataway (2017), 495–4997. 10.1109/ICIP.2017.8296330Search in Google Scholar
[24] D. A. Lorenz and T. Pock, An inertial forward-backward algorithm for monotone inclusions, J. Math. Imaging Vision 51 (2015), no. 2, 311–325. 10.1007/s10851-014-0523-2Search in Google Scholar
[25] J. Markham and J. A. Conchello, Parametric blind deconvolution: A robust method for the simultaneous estimation of image and blur, J. Optical Soc. Amer. A 16 (1999), no. 10, 2377–2391. 10.1364/JOSAA.16.002377Search in Google Scholar PubMed
[26] J. A. P. Oliveira, M. A. T. Figueiredo and J. M. Bioucas-Dias, Parametric blur estimation for blind restoration of natural images: Linear motion and out-of-focus, IEEE Trans. Image Process. 23 (2014), no. 1, 466–477. 10.1109/TIP.2013.2286328Search in Google Scholar PubMed
[27] Z. A. Opial, Weak convergence of the sequence of successive approximations for nonexpansive mappings, Bull. Amer. Math. Soc. 73 (1967), 591–597. 10.1090/S0002-9904-1967-11761-0Search in Google Scholar
[28]
H. Pan and T. Blu,
An iterative linear expansion of thresholds for
[29] D. Perrone and P. Favaro, A logarithmic image prior for blind deconvolution, Int. J. Comput. Vis. 117 (2016), no. 2, 159–172. 10.1007/s11263-015-0857-2Search in Google Scholar
[30] S. Ramani, Z. Liu, J. Rosen, J.-F. Nielsen and J. A. Fessler, Regularization parameter selection for nonlinear iterative image restoration and MRI reconstruction using GCV and SURE-based methods, IEEE Trans. Image Process. 21 (2012), no. 8, 3659–3672. 10.1109/TIP.2012.2195015Search in Google Scholar PubMed PubMed Central
[31] S. J. Reeves and R. M. Mersereau, Blur identification by the method of generalized cross-validation, IEEE Trans. Image Process. 1 (1992), no. 3, 301–311. 10.1109/83.148604Search in Google Scholar PubMed
[32] M. R. Renu, S. Chaudhuri and R. Velmurugan, Convergence analysis of a quadratic upper bounded TV regularizer based blind deconvolution, Signal Process. 106 (2015), no. 6, 174–183. 10.1016/j.sigpro.2014.06.029Search in Google Scholar
[33] P. Ruiz, X. Zhou, J. Mateos, R. Molina and A. K. Katsaggelos, Variational Bayesian blind image deconvolution: A review, Digit. Signal Process. 47 (2015), 116–127. 10.1016/j.dsp.2015.04.012Search in Google Scholar
[34] P. Sarder and A. Nehorai, Deconvolution methods for 3-D fluorescence microscopy images, IEEE Signal Process. Mag. 23 (2006), no. 3, 32–45. 10.1109/MSP.2006.1628876Search in Google Scholar
[35] C. Song, K. Ma, A. Li, X. Chen and X. Xu, Diffraction-limited image reconstruction with SURE for atmospheric turbulence removal, Infrared Phys. Technol. 71 (2015), no. 5, 171–174. 10.1016/j.infrared.2015.03.009Search in Google Scholar
[36] J. L. Starck and E. Pantin, Deconvolution in astronomy: A review, Publ. Astronoical Soc. Pacific 114 (2002), no. 800, 1051–1069. 10.1086/342606Search in Google Scholar
[37] C. M. Stein, Estimation of the mean of a multivariate normal distribution, Ann. Statist. 9 (1981), no. 6, 1135–1151. 10.1214/aos/1176345632Search in Google Scholar
[38] R. J. Tibshirani and J. Taylor, Degrees of freedom in lasso problems, Ann. Statist. 40 (2012), no. 2, 1198–1232. 10.1214/12-AOS1003Search in Google Scholar
[39] B. M. Williams, B. Al-Bander, H. Pratt, S. Lawman, Y. Zhao, Y. Zheng and Y. Shen, Fast blur detection and parametric deconvolution of retinal fundus images, Fetal, Infant and Ophthalmic Medical Image Analysis, Lecture Notes in Comput. Sci. 10554, Springer, Cham (2017), 194–201. 10.1007/978-3-319-67561-9_22Search in Google Scholar
[40] F. Xue and T. Blu, A novel SURE-based criterion for parametric PSF estimation, IEEE Trans. Image Process. 24 (2015), no. 2, 595–607. 10.1109/TIP.2014.2380174Search in Google Scholar PubMed
[41] F. Xue, F. Luisier and T. Blu, Multi-Wiener SURE-LET deconvolution, IEEE Trans. Image Process. 22 (2013), no. 5, 1954–1968. 10.1109/TIP.2013.2240004Search in Google Scholar PubMed
[42] F. Xue and A. G. Yagola, Analysis of point-target detection performance based on ATF and TSF, Infrared Phys. Technol. 52 (2009), no. 5, 166–173. 10.1016/j.infrared.2009.07.002Search in Google Scholar
[43] F. Xue, A. G. Yagola, J. Liu and G. Meng, Recursive SURE for iterative reweighted least square algorithms, Inverse Probl. Sci. Eng. 24 (2016), no. 4, 625–646. 10.1080/17415977.2015.1054822Search in Google Scholar
[44] X. Zhou, M. Vega, F. Zhou, R. Molina and A. Katsaggelos, Fast Bayesian blind deconvolution with Huber super Gaussian priors, Digit. Signal Process. 60 (2017), 122–133. 10.1016/j.dsp.2016.08.008Search in Google Scholar
[45] H. Zou, T. Hastie and R. Tibshirani, On the “degrees of freedom” of the lasso, Ann. Statist. 35 (2007), no. 5, 2173–2192. 10.1214/009053607000000127Search in Google Scholar
© 2020 Walter de Gruyter GmbH, Berlin/Boston
