Abstract
A new two-dimensional (2D) sample-based conjugate gradient (SCG) algorithm is developed for adaptive filtering. This algorithm is based on the conjugate gradient method of optimization and therefore has a fast convergence characteristic. The SCG is computationally simpler than the recursive least squares (RLS) algorithm. The SCG algorithm with the equation-error and output-error methods is investigated for application in image restoration. Simulation results show that the new algorithm significantly outperforms existing algorithms in the restoration of noisy images.
Similar content being viewed by others
References
M. R. Azimi-Sadjadi and H. Pan, Two-dimensional block diagonal LMS adaptive filtering,IEEE Trans. on Signal Processing, 2420–2429, Sept. 1994.
P. H. Bauer, Finite wordlength effects in m-D digital filters with singularities on the stability boundary,IEEE Trans. Signal Processing, vol. SP-40, 894–900, April 1992.
P. H. Bauer and E. I. Jury, Stability analysis of multidimensional (m-D) direct realization of digital filters under the influence of nonlinearities,IEEE Trans. Acous. Speech, and Signal Process., vol. ASP-36, 1770–1780, Nov. 1988.
G. K. Boray and M. D. Srinath, Conjugate gradient techniques for adaptive filtering,IEEE Trans, Circuits and Syst., vol. CAS-39, 1–10, Jan. 1992.
T. Bose and M. Q. Chen, Conjugate gradient method in adaptive bilinear filtering,IEEE Trans. Signal Processing, vol. SP-43, 1503–1508, June 1995.
T. Bose and D. A. Trautman, Stability of first-order multidimensional digital filters under quantization,IEEE Trans. Signal Processing, vol. SP-41, 387–390, Jan. 1993.
S. G. Chen and Y. A. Kao, A new efficient 2-D LMS adaptive filtering algorithm,Proc. Intl. Symp. on Circuits and Systems, 233–236, May 1994.
Y. S. Chung and M. Kanefsky, On 2-D recursive LMS algorithms using ARMA prediction for ADPCM encoding of images,IEEE Trans, on Image Processing, vol. 1, no. 3, July 1992.
P. M. Clarkson,Optimal and Adaptive Signal Processing, CRC Press, Boca Raton, FL, 1993.
P. L. Feintuch, An adaptive recursive LMS filter,Proc. IEEE, vol. 64, 1622–1624, 1976.
M. M. Hadhoud and D. W. Thomas, The two-dimensional adaptive LMS (TDLMS) algorithm,IEEE Trans. Circuits and Syst., vol. CAS-35, 485–494, May 1988.
S. Haykin,Adaptive Filter Theory, Prentice-Hall, Englewood Cliffs, NJ, 1986.
A. Jenning, Influence of the eigenvalue spectrum on the convergence rate of the conjugate gradient method,J. Inst. Math. Appl., vol. 20, 67–72, 1977.
K. S. Joo, T. Bose, and J. W. Hall, Jr., 2-D nonlinear adaptive filtering for image restoration,Proc. IEEE Workshop on Nonlinear Signal and Image Processing, 1995.
J. N. Lin, X. Nie, and R. Unbehauen, Two-dimensional LMS adaptive filter incorporating a local-mean estimator for image processing,IEEE Trans. Circuits Sys.-II: Analog and Digital Signal Processing, vol. CAS-40, 417–428, July 1993.
D. G. Luenberger,Introduction to Linear andNonlinear Programming, Addison-Wesley, Reading, MA, 1973.
W. B. Mikhail and S. M. Ghosh, Two-dimensional variable-step-size sequential adaptive gradient algorithms with applications,IEEE Trans. Circuits and Syst., vol. CAS-38, 1577–1580, Dec. 1991.
T. Soni, B. D. Rao, J. R. Zeidler, and W. H. Ku, Enhancement of images using the 2-D LMS adaptive algorithm,Proc. IEEE Intl. Conf. Acous., Speech, and Signal Processing, 3029–3032, May 1991.
J. C. Strait and W. K. Jenkins, A fast two-dimensional quasi-Newton adaptive filter. Proc. Intl. Symp. on Circuits and Systems, 149–152, May 1994.
H. Youlal, M. Janati-I, and M. Najim, Two-dimensional joint process lattice for adaptive restoration of images,IEEE Trans. Image Processing, 366–378, July 1992.
Author information
Authors and Affiliations
Additional information
This work was supported in part by a grant from the Colorado Advanced Software Institute.
Rights and permissions
About this article
Cite this article
Joo, K.S., Bose, T. & Xu, G.F. Image restoration using a conjugate gradient-based adaptive filtering algorithm. Circuits Systems and Signal Process 16, 197–206 (1997). https://doi.org/10.1007/BF01183274
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF01183274