Mathematics and Computer Science |
|
|
|
|
Nonconvex nonsmooth variational model for Poisson noise removal of gray image |
Yuanpeng ZHANG(),Hongtao CHEN,Weina WANG() |
School of Sciences,Hangzhou Dianzi University,Hangzhou 310018,China |
|
|
Abstract Based on the advantages of nonconvex variational models on image edge-preserving and contrast-preserving, this paper introduces a new nonconvex and nonsmooth variational model together with a fast algorithm for the Poisson noise removal. The proposed model consists of a regularization term and a data fidelity term. The regularization term is formulated by a nonconvex Lipschitz potential function composed of the first-order derivative of images, while the data fitting term is depicted by the nonlinear Kullback-Leibler divergence. By using the proximal linearization strategy, the proposed nonconvex and nonsmooth model can be converted into a series of convex models, which are able to be solved by alternating direction method of multipliers. Moreover, we can also prove the monotonic decreasing property of the objective function value sequence. Numerical experiments show that our model with the proposed algorithm is effective for eliminating Poisson noise and obtains higher SNR values compared to classical methods.
|
Received: 19 November 2021
Published: 21 March 2023
|
|
Corresponding Authors:
Weina WANG
E-mail: 2028251625@qq.com;wnwang@hdu.edu.cn
|
|
Cite this article:
Yuanpeng ZHANG, Hongtao CHEN, Weina WANG. Nonconvex nonsmooth variational model for Poisson noise removal of gray image. Journal of Zhejiang University (Science Edition), 2023, 50(2): 160-166.
URL:
https://www.zjujournals.com/sci/EN/Y2023/V50/I2/160
|
基于非凸非光滑变分模型的灰度图像泊松噪声移除算法
基于非凸变分方法在图像边界结构保持和对比度保持上的优势,针对泊松噪声的移除问题提出一种新的非凸非光滑正则化模型及快速求解算法。模型由非凸Lipschitz势函数复合图像梯度信息的正则化项和非线性Kullback-Leibler数据保真项两部分构成。通过使用临近点线性化策略,将求解非凸变分模型转化为求解一系列凸变分模型,进而使用交替方向乘子法求解。同时证明了算法的目标函数值序列具有单调下降性。实验结果表明,该方法能有效消除图像中的泊松噪声,且信噪比较经典算法有明显提升。
关键词:
泊松噪声移除,
非凸非光滑,
临近点线性化,
交替方向乘子法
|
|
[1] |
WU C L, ZHANG J Y, TAI X C. Augmented Lagrangian method for total variation restoration with non-quadratic fidelity[J]. Inverse Problems and Imaging, 2011, 5(1): 237-261. DOI:10.3934/IPI.2011.5.237
doi: 10.3934/IPI.2011.5.237
|
|
|
[2] |
GAO Y M, LIU F, YANG X P. Total generalized variation restoration with non-quadratic fidelity[J]. Multidimensional Systems and Signal Processing, 2018, 29(4): 1459-1484. DOI:10.1007/s11045-017-0512-x
doi: 10.1007/s11045-017-0512-x
|
|
|
[3] |
RUDIN L I, OSHER S, FATEMI E. Nonlinear total variation based noise removal algorithms[J]. Physica D: Nonlinear Phenomena, 1992, 60(1-4): 259-268. DOI:10.1016/0167-2789(92)90242-F
doi: 10.1016/0167-2789(92)90242-F
|
|
|
[4] |
HE B S, YUAN X M. On the O(1/n) convergence rate of the Douglas-Rachford alternating direction method[J]. SIAM Journal on Numerical Analysis, 2012, 50(2): 700-709. DOI:10.1137/110836936
doi: 10.1137/110836936
|
|
|
[5] |
JIANG L, HUANG J, LYU X G, et al. Alternating direction method for the high-order total variation-based Poisson noise removal problem[J]. Numerical Algorithms, 2015, 69(3): 495-516. 10.1007/s11075-014-9908-y
doi: 10.1007/s11075-014-9908-y
|
|
|
[6] |
HUANG J, HUANG T Z. A nonstationary accelerating alternating direction method for frame-based Poissonian image deblurring[J]. Journal of Computational and Applied Mathematics, 2019, 352(1): 181-193. DOI:10.1016/j.cam.2018.11.028
doi: 10.1016/j.cam.2018.11.028
|
|
|
[7] |
RAHMAN CHOWDHURY M, ZHANG J, QIN J, et al. Poisson image denoising based on fractional-order total variation[J]. Inverse Problems and Imaging, 2020, 14(1): 77-96. DOI:10.3934/ipi.2019064
doi: 10.3934/ipi.2019064
|
|
|
[8] |
NIKOLOVA M. Analysis of the recovery of edges in images and signals by minimizing nonconvex regularized least-squares[J]. Multiscale Modeling & Simulation, 2005, 4(3): 960-991. DOI:10.1137/040619582
doi: 10.1137/040619582
|
|
|
[9] |
SHEN Y, LI S. Restricted p-isometry property and its application for nonconvex compressive sensing[J]. Advances in Computational Mathematics, 2012, 37(3): 441-452. DOI:10.1007/s10444-011-9219-y
doi: 10.1007/s10444-011-9219-y
|
|
|
[10] |
CHEN X J, NG M K, ZHANG C. Non-Lipschitz ℓp -regularization and box constrained model for image restoration[J]. IEEE Transactions on Image Processing, 2012, 21(12): 4709-4721. DOI:10. 1109/TIP.2012.2214051
doi: 10. 1109/TIP.2012.2214051
|
|
|
[11] |
SHEN Y, HAN B, BRAVERMAN E. Adaptive frame-based color image denoising[J]. Applied and Computational Harmonic Analysis, 2016, 41(1): 54-74. DOI:10.1016/j.acha.2015.04.001
doi: 10.1016/j.acha.2015.04.001
|
|
|
[12] |
BAO C L, DONG B, HOU L K, et al. Image restoration by minimizing zero norm of wavelet frame coefficients[J]. Inverse Problems, 2016, 32(11): 115004. DOI:10.1088/0266-5611/32/11/115004
doi: 10.1088/0266-5611/32/11/115004
|
|
|
[13] |
ZENG C, WU C L. On the edge recovery property of noncovex nonsmooth regularization in image restoration[J]. SIAM Journal on Numerical Analysis, 2018, 56(2): 1168-1182. DOI:10.1137/17M1123687
doi: 10.1137/17M1123687
|
|
|
[14] |
NIKOLOVA M, NG M K, ZHANG S Q, et al. Efficient reconstruction of piecewise constant images using nonsmooth nonconvex minimization[J]. SIAM Journal on Imaging Sciences, 2008, 1(1): 2-25. DOI:10.1137/070692285
doi: 10.1137/070692285
|
|
|
[15] |
GAO Y M, WU C L. On a general smoothly truncated regularization for variational piecewise constant image restoration: Construction and convergent algorithms[J]. Inverse Problems, 2020, 36(4): 045007. DOI:10.1088/1361-6420/ab661
doi: 10.1088/1361-6420/ab661
|
|
|
[16] |
WANG W N, WU C L, TAI X C. A globally convergent algorithm for a constrained non-Lipschitz image restoration model[J]. Journal of Scientific Computing, 2020, 83(1): 1-29. DOI:10.1007/s10915-020-01190-4
doi: 10.1007/s10915-020-01190-4
|
|
|
[17] |
CHEN X J, NIU L F, YUAN Y X. Optimality conditions and a smoothing trust region Newton method for non-Lipschitz optimization[J]. SIAM Journal on Optimization, 2013, 23(3): 1528-1552. DOI:10.1137/120871390
doi: 10.1137/120871390
|
|
|
[18] |
HINTERMULLER M, WU T. Nonconvex TV q -models in image restoration: Analysis and a trust-region regularization-based superlinearly convergent solver[J]. SIAM Journal on Imaging Sciences, 2003, 6(3): 1385-1415. DOI:10.1137/110854746
doi: 10.1137/110854746
|
|
|
[19] |
BIAN W, CHEN X J. Linearly constrained non-Lipschitz optimization for image restoration[J]. SIAM Journal on Imaging Sciences, 2015, 8(4): 2294-2322. DOI:10.1137/140985639
doi: 10.1137/140985639
|
|
|
[20] |
LAI M J, XU Y Y, YIN W T. Improved iteratively reweighted least squares for unconstrained smoothed ℓ q minimization[J]. SIAM Journal on Numerical Analysis, 2013, 51(2): 927-957. DOI:10.1137/110840364
doi: 10.1137/110840364
|
|
|
[21] |
CHEN X J, ZHOU W J. Convergence of the reweighted ℓ1 minimization algorithm for ℓ2-ℓ p minimization[J]. Computational Optimization and Applications, 2014, 59(1): 47-61. DOI:10.1007/s10589-013-9553-8
doi: 10.1007/s10589-013-9553-8
|
|
|
[22] |
ZENG C, JIA R, WU C L. An iterative support shrinking algorithm for non-Lipschitz optimization in image restoration[J]. Journal of Mathematical Imaging and Vision, 2019, 61(1): 122-139. DOI:10.1007/s10851-018-0830-0
doi: 10.1007/s10851-018-0830-0
|
|
|
[23] |
ZHENG Z, NG M, WU C L. A globally convergent algorithm for a class of gradient compounded non-Lipschitz models applied to non-additive noise removal[J]. Inverse Problems, 2020, 36(12): 125017. DOI:10.1088/1361-6420/abc793
doi: 10.1088/1361-6420/abc793
|
|
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|