
改进的迭代算法在图像恢复正则化模型中的应用
The Application of Improved Iterative Algorithm to Regularization Model of Image Restoration
根据图像成像过程容易受泊松噪声的影响,提出用Kullback-Leibler距离描述保真项,用平方根复合函数描述正则项,建立具有自适应权系数的能量泛函正则化模型.由于模型的梯度退化和海森矩阵的规模较大,使得无法应用牛顿迭代算法.本文利用退化梯度幅值作为约束集,建立可对角化和容易求逆的海森矩阵,提出改进的牛顿投影迭代算法.仿真表明,该方法取得较小的相对误差、偏差,较高的信噪比和良好的视觉效果.
According to the imaging process is easily affected by Poisson noise, the image restoration regularization model that fidelity term is described by Kullback-Leibler Euclidean and the regularization term is established by the square root compound function, with adaptive weight coefficients, is proposed.For the gradient degeneration and the large scale Hessian matrix, it is unable to apply the Newton iterative algorithm to the model.In this paper, constraint set is introduced by the magnitude value of degeneration gradient, the diagonal and easily computed Hessian matrix is established, and the improved Newton iterative projection algorithm is proposed.Simulation results show the proposed can effectively restore image, such as the lower relative error and deviation, the higher peak signal to noise ratio, and better visual effect.
正则化 / 图像恢复 / 海森矩阵 / 活跃集 {{custom_keyword}} /
regularization / image restoration / Hessian matrix / active set {{custom_keyword}} /
[1] Vogel Curtis R.Computational Methods for Inverse Problems[M].Philadelphia, Pennsylvania:Society for Industrial and Applied Mathematics, 2002.1-183.
[2] 唐利明, 黄大荣.变分框架下的多尺度图像恢复与重建[J].电子学报, 2013, 41(12):2353-2360. Tang Li-ming, Huang Da-rong.Multiscale image restoration and reconstruction in the framework of variation[J].Acta Electronica Sinica, 2013, 41(12):2353-2360.(in Chinese)
[3] Aubert G, Kornprobst Pierre.Mathematical Problems in Image Processing, Partial Differential Equations and the Calculus of Variations[M].New York, USA:Springer-Verlag, 2006.1-371.
[4] 童基均, 刘进, 蔡强.基于全变差的加全最小二乘法PET图像重建[J].电子学报, 2013, 41(4):787-790. Tong Ji Jun, Liu Jin, Cai Qiang.The PET image reconstruction based on weighted least squares and TV penalty[J].Acta Electronica Sinica, 2013, 41(4):787-790.(in Chinese)
[5] Liu X W, Huang L H, Guo Z Y.Adaptive fourth-order partial differential equation filter for image denoising[J].Applied Mathematics Letters, 2011, 24(8):1282-1288.
[6] Dykes L, Reichel L.Simplified GSVD computations for the solution of linear discrete ill-posed problems[J].Journal of Computational and Applied Mathematics, 2014, 255(1):15-27.
[7] Beck Amir, Teboulle Marc.A fast dual proximal gradient algorithm for convex minimization and applications[J].Operations Research Letters, 2014, 42(1):1-6.
[8] Duran Joan, Coll Bartomeu, Sbert Catalina.Chambolle's projection algorithm for total variation denoising[J].Image Processing on Line, 2013, 2013(3):301-321.
[9] Dai Y H, Kou C X.A nonlinear conjugate gradient algorithm with an optimal property and an improved wolfe line search[J].SIAM Journal on Optimization, 2013, 23(1):296-320.
[10] Sun Wenyu, Yuan Yaxiang.Optimization Theory and Methods Nonlinear Programming[M].New York, USA:Springer Science Business Media, 2006, 1-687.
[11] Bai zhengjian, Donatelli Marco, Capizzano Stefano Serra.Fast preconditioners for total variation deblurring with anti-reflective boundary conditions[J].SIAM Journal on Matrix Analysis and Applications, 2011, 32(3):785-805.
[12] J.E.Dennis, Robert B.Schnabel.Numerical Methods for Unconstrained Optimization and Nonlinear Equations[M].Philadelphia:SIAM, 1996.1-378.
[13] Beck Amir, Teboulle Marc.A fast iterative shrinkage-thresholding for linear inverse problems[J].SIAM Journal on Imaging Science, 2009, 2(1):183-202.
[14] Landi G, Piccolomini E Loli.An efficient method for nonnegatively constrained total variation-based of medical images corrupted by poisson noise[J].Computerized Medical Imaging and Graphics, 2012, 36(1):38-46.
[15] Lee Katrina Palmer, Nagy James G, Perrone Lisa.Iterative methods for image deblurring:a matlab object-oriented approach[J].Numerical Algorithms, 2004, 36(1):73-93.
国家自然科学基金 (No.61104211); 江苏省高校自然科学基金 (No.10KJB120004); 江苏师范大学博士人才基金 (No.10XLR27)
/
〈 |
|
〉 |