Abstract
Restoration of images with translation invariant filters always involves a tradeoff between detail preservation and noise removal. The tradeoff which is best is that which produces an image that a human being would judge to be closest to the original image. In this paper, we show how such an optimal tradeoff can be achieved by combining a filter design algorithm with a nonlinear model of the human visual system.
Specifically, we provide a method for combining the Visible Differences Predictor (VDP) with a fast, MAE-based algorithm for the design of stack filters. It uses the VDP to tune a Weighted Mean Absolute Error (WMAE) criterion while this criterion is used to design a stack filter. The results are: (a) a WMAE criterion that is best within a specific class in terms of visual fidelity; (b) a stack filter which is optimal among all stack filters under this criterion; and (c) restored images which are visibly superior to those produced by the original MAE-based stack filter approach.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
P. D. Wendt, E. J. Coyle, and Jr. N. C. Gallagher. Stack filters. IEEE Trans. Acoustics, Speech, and Signal Proc., 34:898–911, August 1986.
P A. Maragos and R. W. Schafer. Morphological fiters — Part II: Their relations to median, order statistic and stack filters. IEEE Trans. Acoust., Speech, Signal Processing, ASSP-35:1153–1184, August 1987.
J.-H. Lin and E.J. Coyle. Minimum mean absolute error estimation over the class of generalized stack filters. IEEE Trans. Acoustics, Speech, and Signal Proc., 38:663–678, April 1990.
O. Yli-Harja, J. Astola, and Y. Neuvo. Analysis of the properties of median and weighted median filters using threshold logic and stack filter representation. IEEE transactions on Signal Processing, SP-39(2):395–410, February 1991.
O. Yli-Harja. Formula for the joint distribution of stack filters. IEEE Signal Processing Letters, 1(9):129–130, September 1994.
Ioan Tăbuş, Doina Petrescu, and Moncef Gabbouj. A training framework for stack boolean filtering — fast optimal design procedures and robustness case study. IEEE Tran. on Image Processing, 5(6):809–826, June 1996.
J. P. Fitch, E. J. Coyle, and Jr. N. C. Gallagher. Median filtering by threshold decomposition. IEEE Trans. Acoustics, Speech, and Signal Proc, 32:1183–1188, December 1984.
J. P. Fitch, E. J. Coyle, and Jr. N. C. Gallagher. Threshold decomposition of multidimensional rank order operators. IEEE Trans. Circuits Syst., 32:445–450, May 1985.
J.-H. Lin and Y.-T. Kim. Fast algorithms for training stack filters. IEEE Trans. Signal Processing, 42(4):772–781, April 1994.
J. Yoo, K. L. Fong, J.-J. Huang, E. J. Coyle, and G. B. Adam III. A fast algorithm for designing stack filters. Accepted by the IEEE Trans. on Image Processing.
S. Daly. The visible difference predictor: an algorithm for the assessment of image fidelity. Chapter 14 in Digital Images and Human Vision, ed. by A. B. Watson, MIT Press, 1993.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer-Verlag London Limited
About this paper
Cite this paper
Coyle, E.J., Huang, JJ. (1998). A Perception-Based Error Criterion and the Design of Stack Filters for Image Restoration. In: Marshall, S., Harvey, N.R., Shah, D. (eds) Noblesse Workshop on Non-Linear Model Based Image Analysis. Springer, London. https://doi.org/10.1007/978-1-4471-1597-7_10
Download citation
DOI: https://doi.org/10.1007/978-1-4471-1597-7_10
Publisher Name: Springer, London
Print ISBN: 978-3-540-76258-4
Online ISBN: 978-1-4471-1597-7
eBook Packages: Springer Book Archive