Abstract
A weighted order statistic filter (WOSF) generates a Boolean function by the specified parameters. However, WOSF with different set of parameters may generate the same Boolean function. In this paper, one proposes an alternative representation for WOS filters to characterize WOSF. This method is based on the maximal margin classification of SVM. From a truth table generated by a linearly separable Boolean function, one takes the inputs and outputs to form a training set. By SVM, it generates a maximal margin hyperplane. The hyperplane has an optimal normal vector and optimal bias, and defines a discriminant function. The discriminant function decides the category of training data and can be used to represent WOSF. In other words, one merely utilizes a normal vector and bias to represent all WOS filters with same output but different weight vectors and threshold values.
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
Wentd, P.D., Coyle, E.J., Gallagher Jr., N.C.: Stack Filters. IEEE Trans. Acoust. Speech and Signal Processing 34(4), 898–911 (1986)
Yli-Harja, O., Astola, J.T., Neuvo, Y.: Analysis of the properties of median and weighted median filters using threshold logic and stack filter representation. IEEE Transactions on Signal Processing 39, 395–410 (1991)
Nieweglowski, J., Gabbouj, M., Neuvo, Y.: Weighted medias-positive Boolean functions conversion algorithms. Signal Processing 34, 149–161 (1982)
Wentd, P.D.: Nonrecursive and recursive stack filters and their filtering behavior. IEEE Trans. Acoust. Speech and Signal Processing 38(12), 2099–2107 (1990)
Koivisto, P., Huttunen, H.: Design Of Weighted Order Statistic Filters By Training-Based Optimization. In: International Symposium on Signal Processing and its Application (ISSPA), Kuala Lumpur, Malaysia (2001)
Ropert, M., de Saint-Martin, F.M., Pele, D.: A new representation of weighted order statistic filters. Signal Processing 54, 201–206 (1996)
Savin, C.E., Ahmad, M.O., Swamy, M.N.S.: Design of Weighted Order Statistic Filters Using Linearly Separable Stack-Like Architecture. In: Proceedings of the 37th Midwest Symposium Circuits and Systems, vol. 2, pp. 753–756 (1994)
Hirata, N.S.T., Hirata Jr., R.: Design of Order Statistic Filters from Examples. In: Proceedings of the XVI Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI 2003), Brazilian, pp. 24–26 (2003)
Lin, Y.-L., Teng, W.-C., Jeng, J.-H., Hsieh, J.-G.: Characterization of Canonical Robust Template Values for a Class of Uncoupled CNNs Implementing Linearly Separable Boolean Functions. WSEAS Transactions on Information Science and Applications 7(2), 940–944 (2005)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chen, W.C., Jeng, J.H. (2006). A Representation of WOS Filters Through Support Vector Machine. In: Chang, LW., Lie, WN. (eds) Advances in Image and Video Technology. PSIVT 2006. Lecture Notes in Computer Science, vol 4319. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11949534_118
Download citation
DOI: https://doi.org/10.1007/11949534_118
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-68297-4
Online ISBN: 978-3-540-68298-1
eBook Packages: Computer ScienceComputer Science (R0)