A comparison of parameter choice rules for (cid:2) p - (cid:2) q minimization

Images that have been contaminated by various kinds of blur and noise can be restored by the minimization of an (cid:2) p - (cid:2) q functional. The quality of the reconstruction depends on the choice of a regularization parameter. Several approaches to determine this parameter have been described in the literature. This work presents a numerical comparison of known approaches as well as of a new one.

[1]  F. Sgallari,et al.  Automatic fidelity and regularization terms selection in variational image restoration , 2021, BIT Numerical Mathematics.

[2]  A. Buccini.  Generalized Cross Validation stopping rule for Iterated Tikhonov regularization , 2021, 2021 21st International Conference on Computational Science and Its Applications (ICCSA).

[3]  Fiorella Sgallari,et al.  ADMM-based residual whiteness principle for automatic parameter selection in super-resolution problems , 2021, ArXiv.

[4]  L. Reichel,et al.  Generalized cross validation for ℓp-ℓq minimization , 2021, Numerical Algorithms.

[5]  Marco Donatelli,et al.  Graph Laplacian for image deblurring , 2021, ETNA - Electronic Transactions on Numerical Analysis.

[6]  Alessandro Buccini,et al.  An ℓp-ℓq minimization method with cross-validation for the restoration of impulse noise contaminated images , 2020, J. Comput. Appl. Math..

[7]  L. Reichel,et al.  Modulus-based iterative methods for constrained ℓp–ℓq minimization , 2020, Inverse Problems.

[8]  Alessandro Buccini,et al.  Generalized singular value decomposition with iterated Tikhonov regularization , 2020, J. Comput. Appl. Math..

[9]  Omar De la Cruz Cabrera,et al.  Large-scale regression with non-convex loss and penalty , 2020, Applied Numerical Mathematics.

[10]  F. Sgallari,et al.  Residual whiteness principle for parameter-free image restoration , 2020 .

[11]  L. Reichel,et al.  Generalized Cross Validation for l p-l q Minimization , 2020 .

[12]  F. Sgallari,et al.  Whiteness Constraints in a Unified Variational Framework for Image Restoration , 2018, Journal of Mathematical Imaging and Vision.

[13]  Runyi Yu,et al.  Residual Correlation Regularization Based Image Denoising , 2018, IEEE Signal Processing Letters.

[14]  L. Reichel,et al.  GCV for Tikhonov regularization by partial SVD , 2017 .

[15]  F. Sgallari,et al.  Majorization–minimization generalized Krylov subspace methods for (cid:2) p – (cid:2) q optimization applied to image restoration , 2017 .

[16]  Yann Gousseau,et al.  Penalizing local correlations in the residual improves image denoising performance , 2016, 2016 24th European Signal Processing Conference (EUSIPCO).

[17]  Lothar Reichel,et al.  GCV for Tikhonov regularization via global Golub–Kahan decomposition , 2016, Numer. Linear Algebra Appl..

[18]  Lea Fleischer,et al.  Regularization of Inverse Problems , 1996 .

[19]  Serena Morigi,et al.  A Generalized Krylov Subspace Method for ℓp-ℓq Minimization , 2015, SIAM J. Sci. Comput..

[20]  Fiorella Sgallari,et al.  A generalized Krylov subspace method for l p-l q minimization , 2015 .

[21]  Serena Morigi,et al.  Variational Image Restoration with Constraints on Noise Whiteness , 2014, Journal of Mathematical Imaging and Vision.

[22]  Anthony J. Yezzi,et al.  Variational Image Denoising Based on Autocorrelation Whiteness , 2013, SIAM J. Imaging Sci..

[23]  Mário A. T. Figueiredo,et al.  Parameter Estimation for Blind and Non-Blind Deblurring Using Residual Whiteness Measures , 2013, IEEE Transactions on Image Processing.

[24]  S. Kindermann.  CONVERGENCE ANALYSIS OF MINIMIZATION-BASED NOISE LEVEL-FRE E PARAMETER CHOICE RULES FOR LINEAR ILL-POSED PROBLEMS , 2011 .

[25]  Raymond H. Chan,et al.  Half-Quadratic Algorithm for ℓp - ℓq Problems with Applications to TV-ℓ1 Image Restoration and Compressive Sensing , 2011, Efficient Algorithms for Global Optimization Methods in Computer Vision.

[26]  Lothar Reichel,et al.  A new zero-finder for Tikhonov regularization , 2008 .

[27]  Dianne P. O'Leary,et al.  Residual periodograms for choosing regularization parameters for ill-posed problems , 2008 .

[28]  Jianhong Shen,et al.  Deblurring images: Matrices, spectra, and filtering , 2007, Math. Comput..

[29]  P. Hansen,et al.  Exploiting Residual Information in the Parameter Choice for Discrete Ill-Posed Problems , 2006 .

[30]  R. Snieder.  Inverse Problems in Geophysics , 2001 .

[31]  Alfred K. Louis,et al.  Medical imaging: state of the art and future development , 1992 .

[32]  P. Hansen.  Rank-Deficient and Discrete Ill-Posed Problems: Numerical Aspects of Linear Inversion , 1987 .

[33]  A. Bakushinskii.  Remarks on choosing a regularization parameter using the quasioptimality and ratio criterion , 1985 .

[34]  I. Galligani.  Optimal Numerical Methods for Direct and Inverse Problems in Hydrology , 1982 .

[35]  G. Stewart,et al.  Reorthogonalization and stable algorithms for updating the Gram-Schmidt QR factorization , 1976 .

[36]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .