A Fast Algorithm for Deblurring Models with Neumann Boundary Conditions

Blur removal is an important problem in signal and image processing. The blurring matrices obtained by using the zero boundary condition (corresponding to assuming dark background outside the scene) are Toeplitz matrices for one-dimensional problems and block-Toeplitz--Toeplitz-block matrices for two-dimensional cases. They are computationally intensive to invert especially in the block case. If the periodic boundary condition is used, the matrices become (block) circulant and can be diagonalized by discrete Fourier transform matrices. In this paper, we consider the use of the Neumann boundary condition (corresponding to a reflection of the original scene at the boundary). The resulting matrices are (block) Toeplitz-plus-Hankel matrices. We show that for symmetric blurring functions, these blurring matrices can always be diagonalized by discrete cosine transform matrices. Thus the cost of inversion is significantly lower than that of using the zero or periodic boundary conditions. We also show that the use of the Neumann boundary condition provides an easy way of estimating the regularization parameter when the generalized cross-validation is used. When the blurring function is nonsymmetric, we show that the optimal cosine transform preconditioner of the blurring matrix is equal to the blurring matrix generated by the symmetric part of the blurring function. Numerical results are given to illustrate the efficiency of using the Neumann boundary condition.

[1]  Stephen A. Martucci,et al.  Symmetric convolution and the discrete sine and cosine transforms , 1993, IEEE Trans. Signal Process..

[2]  M. Buckley.  Fast computation of a discretized thin-plate smoothing spline for image data , 1994 .

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

[4]  Raymond H. Chan,et al.  SINE TRANSFORM BASED PRECONDITIONERS FOR SYMMETRIC TOEPLITZ SYSTEMS , 1996 .

[5]  T. Kailath,et al.  Displacement structure approach to discrete-trigonometric-transform based preconditioners of G.Strang type and of T.Chan type , 1996 .

[6]  Aggelos K. Katsaggelos,et al.  Iterative identification and restoration of images , 1990, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[7]  Jorge Herbert de Lira,et al.  Two-Dimensional Signal and Image Processing , 1989 .

[8]  Raymond H. Chan,et al.  Generalization of Strang's Preconditioner with Applications to Toeplitz Least Squares Problems , 1996, Numer. Linear Algebra Appl..

[9]  J. Huisman.  The Netherlands , 1996, The Lancet.

[10]  Raymond H. Chan,et al.  Conjugate Gradient Methods for Toeplitz Systems , 1996, SIAM Rev..

[11]  D. Manolakis,et al.  Fast algorithms for block toeplitz matrices with toeplitz entries , 1984 .

[12]  Antonio M. Peinado,et al.  Diagonalizing properties of the discrete cosine transforms , 1995, IEEE Trans. Signal Process..

[13]  Franklin T. Luk,et al.  Reducing boundary distortion in image restoration , 1994, Optics & Photonics.

[14]  P. Yip,et al.  Discrete Cosine Transform: Algorithms, Advantages, Applications , 1990 .

[15]  N. Wiener.  The Wiener RMS (Root Mean Square) Error Criterion in Filter Design and Prediction , 1949 .

[16]  Raymond H. Chan,et al.  Cosine transform based preconditioners for total variation deblurring , 1999, IEEE Trans. Image Process..

[17]  Anil K. Jain.  Fundamentals of Digital Image Processing , 2018, Control of Color Imaging Systems.

[18]  N. Levinson.  The Wiener (Root Mean Square) Error Criterion in Filter Design and Prediction , 1946 .

[19]  W. Gragg,et al.  Superfast solution of real positive definite toeplitz systems , 1988 .

[20]  Ali H. Sayed,et al.  Displacement Structure: Theory and Applications , 1995, SIAM Rev..