The influence of the regularization parameter and the first estimate on the performance of tikhonov regularized non-linear image restoration algorithms

Summary This paper reports studies on the influence of the regularization parameter and the first estimate on the performance of iterative image restoration algorithms. We discuss regularization parameter estimation methods that have been developed for the linear Tikhonov‐Miller filter to restore images distorted by additive Gaussian noise. We have performed experiments on synthetic data to show that these methods can be used to determine the regularization parameter of non-linear iterative image restoration algorithms, which we use to restore images contaminated by Poisson noise. We conclude that the generalized crossvalidation method is an efficient method to determine a value of the regularization parameter close to the optimal value. We have also derived a method to estimate the regularization parameter of a Tikhonov regularized version of the Richardson‐Lucy algorithm. These iterative image restoration algorithms need a first estimate to start their iteration. An obvious and frequently used choice for the first estimate is the acquired image. However, the restoration algorithm could be sensitive to the noise present in this image, which may hamper the convergence of the algorithm. We have therefore compared various choices of first estimates and tested the convergence of various iterative restoration algorithms. We found that most algorithms converged for most choices, but that smoothed first estimates resulted in a faster convergence.

[1]  Lucas J. van Vliet,et al.  Application of image restoration methods for confocal fluorescence microscopy , 1997, Photonics West - Biomedical Optics.

[2]  William H. Press,et al.  Numerical Recipes in Fortran 77: The Art of Scientific Computing 2nd Editionn - Volume 1 of Fortran Numerical Recipes , 1992 .

[3]  Nikolas P. Galatsanos,et al.  Methods for choosing the regularization parameter and estimating the noise variance in image restoration and their relation , 1992, IEEE Trans. Image Process..

[4]  William H. Richardson,et al.  Bayesian-Based Iterative Method of Image Restoration , 1972 .

[5]  Timothy J. Holmes.  Maximum-likelihood image restoration adapted for noncoherent optical imaging , 1988 .

[6]  K. Miller.  Least Squares Methods for Ill-Posed Problems with a Prescribed Bound , 1970 .

[7]  J. Conchello,et al.  Enhanced three-dimensional reconstruction from confocal scanning microscope images. II. Depth discrimination versus signal-to-noise ratio in partially confocal images. , 1994, Applied optics.

[8]  F S Fay,et al.  Superresolution three-dimensional images of fluorescence in cells with minimal light exposure. , 1995, Science.

[9]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[10]  William H. Press,et al.  Numerical recipes in C , 2002 .

[11]  Donald L. Snyder,et al.  Random Point Processes in Time and Space , 1991 .

[12]  Geert M. P. van Kempen,et al.  Image Restoration in Fluorescence Microscopy , 1999 .

[13]  B. R. Hunt,et al.  Digital Image Restoration , 1977 .

[14]  Linda Kaufman,et al.  Implementing and Accelerating the EM Algorithm for Positron Emission Tomography , 1987, IEEE Transactions on Medical Imaging.

[15]  W. A. Carrington.  Image restoration in 3-D microscopy with limited data , 1990, Photonics West - Lasers and Applications in Science and Engineering.

[16]  L. Shepp,et al.  Maximum Likelihood Reconstruction for Emission Tomography , 1983, IEEE Transactions on Medical Imaging.

[17]  William H. Press,et al.  Numerical Recipes in C, 2nd Edition , 1992 .

[18]  Peter J. Verveer,et al.  Efficient superresolution restoration algorithms using maximum a posteriori estimations with application to fluorescence microscopy , 1997 .

[19]  Russell M. Mersereau,et al.  Blur identification by the method of generalized cross-validation , 1992, IEEE Trans. Image Process..

[20]  R. White,et al.  Image recovery from data acquired with a charge-coupled-device camera. , 1993, Journal of the Optical Society of America. A, Optics and image science.

[21]  G. M. P. VAN KEMPEN,et al.  A quantitative comparison of image restoration methods for confocal microscopy , 1997 .

[22]  H. M. Voort,et al.  Restoration of confocal images for quantitative image analysis , 1995 .

[23]  L. Shepp,et al.  A Statistical Model for Positron Emission Tomography , 1985 .

[24]  W. Reeves.  Royal Microscopical Society , 1873, Nature.

[25]  A. M. Thompson.  On the Use of Quadratic Regularisation within Maximum Entropy Image Restoration , 1989 .

[26]  Florin Popentiu,et al.  Iterative identification and restoration of images , 1993, Comput. Graph..

[27]  Jose-Angel Conchello,et al.  Fast regularization technique for expectation maximization algorithm for optical sectioning microscopy , 1996, Electronic Imaging.

[28]  A. N. Tikhonov,et al.  Solutions of ill-posed problems , 1977 .

[29]  P. Verveer,et al.  A comparison of image restoration approaches applied to three‐dimensional confocal and wide‐field fluorescence microscopy , 1999, Journal of microscopy.