# American Institute of Mathematical Sciences

August  2015, 9(3): 875-894. doi: 10.3934/ipi.2015.9.875

## A reweighted $l^2$ method for image restoration with Poisson and mixed Poisson-Gaussian noise

 1 Department of mathematics, National University of Singapore, Block S17, 10 Lower Kent Ridge Road, 119076, Singapore 2 Dept. of Math., National Univ. of Singapore, 119076 3 Zhiyuan College, Shanghai Jiao Tong University, 800, Dongchuan Road, Shanghai, 200240, China 4 Department of Mathematics, MOE-LSC and Institute of Natural Sciences, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai, 200240

Received  June 2013 Revised  March 2014 Published  July 2015

We study weighted $l^2$ fidelity in variational models for Poisson noise related image restoration problems. Gaussian approximation to Poisson noise statistic is adopted to deduce weighted $l^2$ fidelity. Different from the traditional weighted $l^2$ approximation, we propose a reweighted $l^2$ fidelity with sparse regularization by wavelet frame. Based on the split Bregman algorithm introduced in [21], the proposed numerical scheme is composed of three easy subproblems that involve quadratic minimization, soft shrinkage and matrix vector multiplications. Unlike usual least square approximation of Poisson noise, we dynamically update the underlying noise variance from previous estimate. The solution of the proposed algorithm is shown to be the same as the one obtained by minimizing Kullback-Leibler divergence fidelity with the same regularization. This reweighted $l^2$ formulation can be easily extended to mixed Poisson-Gaussian noise case. Finally, the efficiency and quality of the proposed algorithm compared to other Poisson noise removal methods are demonstrated through denoising and deblurring examples. Moreover, mixed Poisson-Gaussian noise tests are performed on both simulated and real digital images for further illustration of the performance of the proposed method.
Citation: Jia Li, Zuowei Shen, Rujie Yin, Xiaoqun Zhang. A reweighted $l^2$ method for image restoration with Poisson and mixed Poisson-Gaussian noise. Inverse Problems & Imaging, 2015, 9 (3) : 875-894. doi: 10.3934/ipi.2015.9.875
##### References:
 [1] F. J. Anscombe, The transform of poisson, binomial and negative-binomial data, Biometrika, 35 (1948), 246-254. doi: 10.1093/biomet/35.3-4.246.  Google Scholar [2] M. Bertero, P. Boccacci, G. Desidera and G. Vicidomini, Image deblurring with Poisson data: From cells to galaxies, Inverse Problems, 25 (2009), 123006, 26pp. doi: 10.1088/0266-5611/25/12/123006.  Google Scholar [3] M. Bertero, P. Boccacci, G. Talenti, R. Zanella and L. Zanni, A discrepancy principle for Poisson data, Inverse Problems, 26 (2010), 105004, 20pp. doi: 10.1088/0266-5611/26/10/105004.  Google Scholar [4] P. J. Bickel and E. Levina, Regularized estimation of large covariance matrices, Ann. Statistics, 36 (2008), 199-227. doi: 10.1214/009053607000000758.  Google Scholar [5] C. Brune, A. Sawatzky and M. Burger, Bregman-EM-TV methods with application to optical nanoscopy, in Scale Space and Variational Methods in Computer Vision, Lecture Notes in Computer Science, 5567, Springer, Berlin-Heidelberg, 2009, 235-246. doi: 10.1007/978-3-642-02256-2_20.  Google Scholar [6] C. Brune, A. Sawatzky and M. Burger, Primal and dual Bregman methods with application to optical nanoscopy, Int. J. Comput. Vis., 92 (2011), 211-229. doi: 10.1007/s11263-010-0339-5.  Google Scholar [7] C. Brune, M. Burger, A. Sawatzky, T. Kösters and Frank Wübberling, Forward-Backward EM-TV methods for inverse problems with Poisson noise, preprint, 2009. Google Scholar [8] J.-F. Cai, S. Osher and Z. Shen, Linearized Bregman iterations for compressed sensing, Mathematics of Computation, 78 (2009), 1515-1536. doi: 10.1090/S0025-5718-08-02189-3.  Google Scholar [9] J.-F. Cai, S. Osher and Z. Shen, Split bregman methods and frame based image restoration,, Multiscale Modeling & Simulation, 8 (): 337.  doi: 10.1137/090753504.  Google Scholar [10] J.-F. Cai, B. Dong, S. Osher and Z. Shen, Image restoration: Total variation; wavelet frames; and beyond, J. Amer. Math. Soc., 25 (2012), 1033-1089. doi: 10.1090/S0894-0347-2012-00740-1.  Google Scholar [11] I. Csiszár, Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems, Ann. Statist., 19 (1991), 2032-2066. doi: 10.1214/aos/1176348385.  Google Scholar [12] I. Daubechies, B. Han, A. Ron and Z. Shen, Framelets: MRA-based constructions of wavelet frames, Applied and Computational Harmonic Analysis, 14 (2003), 1-46. doi: 10.1016/S1063-5203(02)00511-0.  Google Scholar [13] B. Dong and Z. Shen, MRA-based wavelet frames and applications, in Mathematics in Image Processing, IAS/Park City Math. Ser., 19, Amer. Math. Soc., Providence, RI, 2013, 9-158.  Google Scholar [14] B. Dong and Y. Zhang, An efficient algorithm for $l^0$ minimization in wavelet frame based image restoration, Journal of Scientific Computing, 54 (2013), 350-368. doi: 10.1007/s10915-012-9597-4.  Google Scholar [15] D. Donoho, Compressed sensing, IEEE Trans. Inform. Theory, 52 (2006), 1289-1306. doi: 10.1109/TIT.2006.871582.  Google Scholar [16] E. Esser, Applications of Lagrangian-Based Alternating Direction Methods and Connections to Split Bregman, UCLA CAM report 09-31, 2009. Google Scholar [17] E. Esser, X. Zhang and T.-F. Chan, A general framework for a class of first order primal-dual algorithms for convex optimization in imaging science, SIAM J. Imag. Sci., 3 (2010), 1015-1046. doi: 10.1137/09076934X.  Google Scholar [18] M. Figueiredo and J. Bioucas-Dias, Restoration of Poissonian images using alternating direction optimization, IEEE Transactions on Image Processing, 19 (2010), 3133-3145. doi: 10.1109/TIP.2010.2053941.  Google Scholar [19] J. Friedman, T. Hastie and R. Tibshirani, Sparse inverse covariance estimation with the graphical lasso, Biostatistics, 9 (2008), 432-441. doi: 10.1093/biostatistics/kxm045.  Google Scholar [20] T. Goldstein, B. O'Donoghue and S. Setzer, Fast alternating direction optimization methods, SIAM J. Imaging Sci., 7 (2015), 1588-1623. doi: 10.1137/120896219.  Google Scholar [21] T. Goldstein and S. Osher, The split bregman method for $l^1$ regularized problems, SIAM Journal on Imaging Sciences, 2 (2009), 323-343. doi: 10.1137/080725891.  Google Scholar [22] Z. Gong, Z. Shen and K.-C. Toh, Image restoration with mixed or unknown noises, Multiscale Model. Simul., 12 (2014), 458-487. doi: 10.1137/130904533.  Google Scholar [23] K. Lange and R. Carson, EM reconstruction algorithms for emission and transmission tomography, J. Comput. Assist. Tomogr., 8 (1984), 306-316. Google Scholar [24] F. Luisier, T. Blu and M. Unser, Image denoising in mixed Poisson-Gaussian niose, IEEE Trans. Image Process., 20 (2011), 696-708. doi: 10.1109/TIP.2010.2073477.  Google Scholar [25] Y. Nesterov, A method of solving a convex programming problem with convergence rate $o(1 / k^2)$, (Russian) Dokl. Akad. Nauk SSSR, 269 (1983), 543-547.  Google Scholar [26] R. Rockafellar, Augmented lagrangians and applications of the proximal point algorithm in convex programming, Mathematics of Operations Research, 1 (1976), 97-116. doi: 10.1287/moor.1.2.97.  Google Scholar [27] A. Ron and Z. Shen, Affine systems in $L_2(R^d)$: The analysis of the analysis operator, Journal of Functional Analysis, 148 (1997), 408-447. doi: 10.1006/jfan.1996.3079.  Google Scholar [28] L. I. Rudin, S. Osher and E. Fatemi, Nonlinear total variation based noise removal algorithms, Physics D, 60 (1992), 259-268. doi: 10.1016/0167-2789(92)90242-F.  Google Scholar [29] S. Setzer, Split bregman algorithm, douglas-rachford splitting and frame shrinkage, in Scale Space and Variational Methods in Computer Vision, Lecture Notes in Computer Science, 5567, Springer, Berlin-Heidelberg, 2009, 464-476. doi: 10.1007/978-3-642-02256-2_39.  Google Scholar [30] S. Setzer, G. Steidl and T. Teuber, Deblurring Poissonian images by split Bregman techniques, Journal of Visual Communication and Image Representation, 21 (2010), 193-199. doi: 10.1016/j.jvcir.2009.10.006.  Google Scholar [31] L. A. Shepp and Y. Vardi, Maximum Likelihood Reconstruction in Positron Emission Tomography, IEEE Transactions on Medical Imaging, 1 (1982), 113-122. Google Scholar [32] D. L. Snyder, A. M. Hammoud and R. L. White, Image recovery from data acquired with a charge-coupled-device camera, J. Opt. Soc. Am. A, 10 (1993), 1014-1023. doi: 10.1364/JOSAA.10.001014.  Google Scholar [33] A. Staglianò, P. Boccacci and M. Bertero, Analysis of an approximate model for Poisson data reconstruction and a related discrepancy principle, Inverse Problems, 27 (2011), 125003, 20pp. doi: 10.1088/0266-5611/27/12/125003.  Google Scholar [34] C. Tomasi and R. Manduchi, Bilateral filtering for gray and color images, in Computer Vision, 1998. Sixth International Conference on, IEEE, 1998, 839-846. doi: 10.1109/ICCV.1998.710815.  Google Scholar [35] H. Yu and G. Wang, Compressed sensing based interior tomography, Physics in Medicine and Biology, 54 (2009), p2791. Google Scholar [36] X. Zhang, M. Burger and S. Osher, A unified primal-dual algorithm framework based on bregman iteration, Journal of Scientific Computing, 46 (2011), 20-46. doi: 10.1007/s10915-010-9408-8.  Google Scholar

show all references

##### References:
 [1] F. J. Anscombe, The transform of poisson, binomial and negative-binomial data, Biometrika, 35 (1948), 246-254. doi: 10.1093/biomet/35.3-4.246.  Google Scholar [2] M. Bertero, P. Boccacci, G. Desidera and G. Vicidomini, Image deblurring with Poisson data: From cells to galaxies, Inverse Problems, 25 (2009), 123006, 26pp. doi: 10.1088/0266-5611/25/12/123006.  Google Scholar [3] M. Bertero, P. Boccacci, G. Talenti, R. Zanella and L. Zanni, A discrepancy principle for Poisson data, Inverse Problems, 26 (2010), 105004, 20pp. doi: 10.1088/0266-5611/26/10/105004.  Google Scholar [4] P. J. Bickel and E. Levina, Regularized estimation of large covariance matrices, Ann. Statistics, 36 (2008), 199-227. doi: 10.1214/009053607000000758.  Google Scholar [5] C. Brune, A. Sawatzky and M. Burger, Bregman-EM-TV methods with application to optical nanoscopy, in Scale Space and Variational Methods in Computer Vision, Lecture Notes in Computer Science, 5567, Springer, Berlin-Heidelberg, 2009, 235-246. doi: 10.1007/978-3-642-02256-2_20.  Google Scholar [6] C. Brune, A. Sawatzky and M. Burger, Primal and dual Bregman methods with application to optical nanoscopy, Int. J. Comput. Vis., 92 (2011), 211-229. doi: 10.1007/s11263-010-0339-5.  Google Scholar [7] C. Brune, M. Burger, A. Sawatzky, T. Kösters and Frank Wübberling, Forward-Backward EM-TV methods for inverse problems with Poisson noise, preprint, 2009. Google Scholar [8] J.-F. Cai, S. Osher and Z. Shen, Linearized Bregman iterations for compressed sensing, Mathematics of Computation, 78 (2009), 1515-1536. doi: 10.1090/S0025-5718-08-02189-3.  Google Scholar [9] J.-F. Cai, S. Osher and Z. Shen, Split bregman methods and frame based image restoration,, Multiscale Modeling & Simulation, 8 (): 337.  doi: 10.1137/090753504.  Google Scholar [10] J.-F. Cai, B. Dong, S. Osher and Z. Shen, Image restoration: Total variation; wavelet frames; and beyond, J. Amer. Math. Soc., 25 (2012), 1033-1089. doi: 10.1090/S0894-0347-2012-00740-1.  Google Scholar [11] I. Csiszár, Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems, Ann. Statist., 19 (1991), 2032-2066. doi: 10.1214/aos/1176348385.  Google Scholar [12] I. Daubechies, B. Han, A. Ron and Z. Shen, Framelets: MRA-based constructions of wavelet frames, Applied and Computational Harmonic Analysis, 14 (2003), 1-46. doi: 10.1016/S1063-5203(02)00511-0.  Google Scholar [13] B. Dong and Z. Shen, MRA-based wavelet frames and applications, in Mathematics in Image Processing, IAS/Park City Math. Ser., 19, Amer. Math. Soc., Providence, RI, 2013, 9-158.  Google Scholar [14] B. Dong and Y. Zhang, An efficient algorithm for $l^0$ minimization in wavelet frame based image restoration, Journal of Scientific Computing, 54 (2013), 350-368. doi: 10.1007/s10915-012-9597-4.  Google Scholar [15] D. Donoho, Compressed sensing, IEEE Trans. Inform. Theory, 52 (2006), 1289-1306. doi: 10.1109/TIT.2006.871582.  Google Scholar [16] E. Esser, Applications of Lagrangian-Based Alternating Direction Methods and Connections to Split Bregman, UCLA CAM report 09-31, 2009. Google Scholar [17] E. Esser, X. Zhang and T.-F. Chan, A general framework for a class of first order primal-dual algorithms for convex optimization in imaging science, SIAM J. Imag. Sci., 3 (2010), 1015-1046. doi: 10.1137/09076934X.  Google Scholar [18] M. Figueiredo and J. Bioucas-Dias, Restoration of Poissonian images using alternating direction optimization, IEEE Transactions on Image Processing, 19 (2010), 3133-3145. doi: 10.1109/TIP.2010.2053941.  Google Scholar [19] J. Friedman, T. Hastie and R. Tibshirani, Sparse inverse covariance estimation with the graphical lasso, Biostatistics, 9 (2008), 432-441. doi: 10.1093/biostatistics/kxm045.  Google Scholar [20] T. Goldstein, B. O'Donoghue and S. Setzer, Fast alternating direction optimization methods, SIAM J. Imaging Sci., 7 (2015), 1588-1623. doi: 10.1137/120896219.  Google Scholar [21] T. Goldstein and S. Osher, The split bregman method for $l^1$ regularized problems, SIAM Journal on Imaging Sciences, 2 (2009), 323-343. doi: 10.1137/080725891.  Google Scholar [22] Z. Gong, Z. Shen and K.-C. Toh, Image restoration with mixed or unknown noises, Multiscale Model. Simul., 12 (2014), 458-487. doi: 10.1137/130904533.  Google Scholar [23] K. Lange and R. Carson, EM reconstruction algorithms for emission and transmission tomography, J. Comput. Assist. Tomogr., 8 (1984), 306-316. Google Scholar [24] F. Luisier, T. Blu and M. Unser, Image denoising in mixed Poisson-Gaussian niose, IEEE Trans. Image Process., 20 (2011), 696-708. doi: 10.1109/TIP.2010.2073477.  Google Scholar [25] Y. Nesterov, A method of solving a convex programming problem with convergence rate $o(1 / k^2)$, (Russian) Dokl. Akad. Nauk SSSR, 269 (1983), 543-547.  Google Scholar [26] R. Rockafellar, Augmented lagrangians and applications of the proximal point algorithm in convex programming, Mathematics of Operations Research, 1 (1976), 97-116. doi: 10.1287/moor.1.2.97.  Google Scholar [27] A. Ron and Z. Shen, Affine systems in $L_2(R^d)$: The analysis of the analysis operator, Journal of Functional Analysis, 148 (1997), 408-447. doi: 10.1006/jfan.1996.3079.  Google Scholar [28] L. I. Rudin, S. Osher and E. Fatemi, Nonlinear total variation based noise removal algorithms, Physics D, 60 (1992), 259-268. doi: 10.1016/0167-2789(92)90242-F.  Google Scholar [29] S. Setzer, Split bregman algorithm, douglas-rachford splitting and frame shrinkage, in Scale Space and Variational Methods in Computer Vision, Lecture Notes in Computer Science, 5567, Springer, Berlin-Heidelberg, 2009, 464-476. doi: 10.1007/978-3-642-02256-2_39.  Google Scholar [30] S. Setzer, G. Steidl and T. Teuber, Deblurring Poissonian images by split Bregman techniques, Journal of Visual Communication and Image Representation, 21 (2010), 193-199. doi: 10.1016/j.jvcir.2009.10.006.  Google Scholar [31] L. A. Shepp and Y. Vardi, Maximum Likelihood Reconstruction in Positron Emission Tomography, IEEE Transactions on Medical Imaging, 1 (1982), 113-122. Google Scholar [32] D. L. Snyder, A. M. Hammoud and R. L. White, Image recovery from data acquired with a charge-coupled-device camera, J. Opt. Soc. Am. A, 10 (1993), 1014-1023. doi: 10.1364/JOSAA.10.001014.  Google Scholar [33] A. Staglianò, P. Boccacci and M. Bertero, Analysis of an approximate model for Poisson data reconstruction and a related discrepancy principle, Inverse Problems, 27 (2011), 125003, 20pp. doi: 10.1088/0266-5611/27/12/125003.  Google Scholar [34] C. Tomasi and R. Manduchi, Bilateral filtering for gray and color images, in Computer Vision, 1998. Sixth International Conference on, IEEE, 1998, 839-846. doi: 10.1109/ICCV.1998.710815.  Google Scholar [35] H. Yu and G. Wang, Compressed sensing based interior tomography, Physics in Medicine and Biology, 54 (2009), p2791. Google Scholar [36] X. Zhang, M. Burger and S. Osher, A unified primal-dual algorithm framework based on bregman iteration, Journal of Scientific Computing, 46 (2011), 20-46. doi: 10.1007/s10915-010-9408-8.  Google Scholar
 [1] Jie Zhang, Yuping Duan, Yue Lu, Michael K. Ng, Huibin Chang. Bilinear constraint based ADMM for mixed Poisson-Gaussian noise removal. Inverse Problems & Imaging, 2021, 15 (2) : 339-366. doi: 10.3934/ipi.2020071 [2] Mingchao Zhao, You-Wei Wen, Michael Ng, Hongwei Li. A nonlocal low rank model for poisson noise removal. Inverse Problems & Imaging, 2021, 15 (3) : 519-537. doi: 10.3934/ipi.2021003 [3] Jianbin Yang, Cong Wang. A wavelet frame approach for removal of mixed Gaussian and impulse noise on surfaces. Inverse Problems & Imaging, 2017, 11 (5) : 783-798. doi: 10.3934/ipi.2017037 [4] Bruno Sixou, Tom Hohweiller, Nicolas Ducros. Morozov principle for Kullback-Leibler residual term and Poisson noise. Inverse Problems & Imaging, 2018, 12 (3) : 607-634. doi: 10.3934/ipi.2018026 [5] Thorsten Hohage, Mihaela Pricop. Nonlinear Tikhonov regularization in Hilbert scales for inverse boundary value problems with random noise. Inverse Problems & Imaging, 2008, 2 (2) : 271-290. doi: 10.3934/ipi.2008.2.271 [6] Xueqin Li, Chao Tang, Tianmin Huang. Poisson $S^2$-almost automorphy for stochastic processes and its applications to SPDEs driven by Lévy noise. Discrete & Continuous Dynamical Systems - B, 2018, 23 (8) : 3309-3345. doi: 10.3934/dcdsb.2018282 [7] Leonid Shaikhet. Stability of delay differential equations with fading stochastic perturbations of the type of white noise and poisson's jumps. Discrete & Continuous Dynamical Systems - B, 2020, 25 (9) : 3651-3657. doi: 10.3934/dcdsb.2020077 [8] Johnathan M. Bardsley. A theoretical framework for the regularization of Poisson likelihood estimation problems. Inverse Problems & Imaging, 2010, 4 (1) : 11-17. doi: 10.3934/ipi.2010.4.11 [9] Yan Wang, Lei Wang, Yanxiang Zhao, Aimin Song, Yanping Ma. A stochastic model for microbial fermentation process under Gaussian white noise environment. Numerical Algebra, Control & Optimization, 2015, 5 (4) : 381-392. doi: 10.3934/naco.2015.5.381 [10] Boris P. Belinskiy, Peter Caithamer. Energy estimate for the wave equation driven by a fractional Gaussian noise. Conference Publications, 2007, 2007 (Special) : 92-101. doi: 10.3934/proc.2007.2007.92 [11] Boris P. Belinskiy, Peter Caithamer. Energy of an elastic mechanical system driven by Gaussian noise white in time. Conference Publications, 2001, 2001 (Special) : 39-49. doi: 10.3934/proc.2001.2001.39 [12] C. Davini, F. Jourdan. Approximations of degree zero in the Poisson problem. Communications on Pure & Applied Analysis, 2005, 4 (2) : 267-281. doi: 10.3934/cpaa.2005.4.267 [13] Hanwool Na, Myeongmin Kang, Miyoun Jung, Myungjoo Kang. Nonconvex TGV regularization model for multiplicative noise removal with spatially varying parameters. Inverse Problems & Imaging, 2019, 13 (1) : 117-147. doi: 10.3934/ipi.2019007 [14] Yuying Shi, Zijin Liu, Xiaoying Wang, Jinping Zhang. Edge detection with mixed noise based on maximum a posteriori approach. Inverse Problems & Imaging, , () : -. doi: 10.3934/ipi.2021035 [15] Luca Rondi. On the regularization of the inverse conductivity problem with discontinuous conductivities. Inverse Problems & Imaging, 2008, 2 (3) : 397-409. doi: 10.3934/ipi.2008.2.397 [16] Marius Ghergu, Gurpreet Singh. On a class of mixed Choquard-Schrödinger-Poisson systems. Discrete & Continuous Dynamical Systems - S, 2019, 12 (2) : 297-309. doi: 10.3934/dcdss.2019021 [17] Wolfgang Arendt, Daniel Daners. Varying domains: Stability of the Dirichlet and the Poisson problem. Discrete & Continuous Dynamical Systems, 2008, 21 (1) : 21-39. doi: 10.3934/dcds.2008.21.21 [18] Daniel Gerth, Andreas Hofinger, Ronny Ramlau. On the lifting of deterministic convergence rates for inverse problems with stochastic noise. Inverse Problems & Imaging, 2017, 11 (4) : 663-687. doi: 10.3934/ipi.2017031 [19] Alain Bensoussan, Xinwei Feng, Jianhui Huang. Linear-quadratic-Gaussian mean-field-game with partial observation and common noise. Mathematical Control & Related Fields, 2021, 11 (1) : 23-46. doi: 10.3934/mcrf.2020025 [20] Jian-Feng Cai, Raymond H. Chan, Mila Nikolova. Two-phase approach for deblurring images corrupted by impulse plus gaussian noise. Inverse Problems & Imaging, 2008, 2 (2) : 187-204. doi: 10.3934/ipi.2008.2.187

2020 Impact Factor: 1.639