Abstract
Support Vector Machines (SVMs) have been used successfully for many classification tasks. In this paper, we investigate applying SVMs to classification in the context of image processing. We chose to look at classifying whether pixels have been corrupted by impulsive noise, as this is one of the simpler classification tasks in image processing. We found that the straightforward application of SVMs to this problem led to a number of difficulties, such as long training times, performance that was sensitive to the balance of classes in the training data, and poor classification performance overall. We suggest remedies for some of these problems, including the use of image filters to suppress variation in the training data. This led us to develop a two-stage classification process which used SVMs in the second stage. This two-stage process was able to achieve substantially better results than those resulting from the straightforward application of SVMs.
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
Burges, C.: A tutorial on Support Vector Machines for pattern recognition. Data Mining and Knowledge Discovery 2, 121–167 (1998)
Cristianini, N., Shawe-Taylor, J.: An Introduction to Support Vector Machines and other kernel-based learning methods. Cambridge University Press, Cambridge (2000)
Davies, E.R.: Edge location shifts produced by median filters: theoretical bounds and experimental results. Signal Process 16, 83–96 (1989)
Davies, E.R.: A remanent noise problem with the median filter. In: Proc. 11th IAPR International Conference on Image, Speech and Signal Analysis, vol. 3, pp. 505–508 (1992)
El-Naqa, I., Yang, Y., Wernick, M., Galatsanos, N., Nishikawa, R.: A Support Vector Machine approach for detection of microcalcifications. IEEE Transactions on Medical Imaging 21, 1552–1563 (2002)
Hearst, M., Dumais, S., Osuna, E., Platt, J., Schölkopf, B.: Support Vector Machines. IEEE Intelligent Systems 13, 18–28 (1998)
Immerkær, J.: Fast noise variance estimation. Computer vision and image understanding 64, 300–302 (1996)
Joachims, T.: Making large-scale SVM learning practical. In: Schölkopf, B., Burges, C., Smola, A. (eds.) Advances in Kernel Methods - Support Vector Learning, pp. 169–184. MIT Press, Cambridge (1999)
Platt, J.: Chapter 12. In: Schölkopf, B., Burges, C., Smola, A. (eds.) Advances in Kernel Methods - Support Vector Learning, pp. 185–208. MIT Press, Cambridge (1999)
Tong, S., Chang, E.: Support Vector Machine active learning for image retrieval. In: Proceedings of the ninth ACM international conference on Multimedia, pp. 107–118. ACM Press, New York (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Clarke, D., Albrecht, D., Tischer, P. (2004). An Investigation into Applying Support Vector Machines to Pixel Classification in Image Processing. In: Webb, G.I., Yu, X. (eds) AI 2004: Advances in Artificial Intelligence. AI 2004. Lecture Notes in Computer Science(), vol 3339. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30549-1_13
Download citation
DOI: https://doi.org/10.1007/978-3-540-30549-1_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-24059-4
Online ISBN: 978-3-540-30549-1
eBook Packages: Computer ScienceComputer Science (R0)