An Efficient Algorithm for Optimal Multilevel Thresholding of Irregularly Sampled Histograms
Optimal multilevel thresholding is a quite important problem in image segmentation and pattern recognition. Although efficient algorithms have been proposed recently, they do not address the issue of irregularly sampled histograms. A polynomial-time algorithm for multilevel thresholding of irregularly sampled histograms is proposed. The algorithm is polynomial not just on the number of bins of the histogram, n, but also on the number of thresholds, k, i.e. it runs in Θ(kn 2). The proposed algorithm is general enough for a wide range of thresholding and clustering criteria, and has the capability of dealing with irregularly sampled histograms. This implies important consequences on pattern recognition, since optimal clustering in the one-dimensional space can be obtained in polynomial time. Experiments on synthetic and real-life histograms show that for typical cases, the proposed algorithm can find the optimal thresholds in a fraction of a second.
KeywordsParticle Swarm Optimization Optimal Threshold Optimal Cluster Pattern Recognition Letter Algorithm Multilevel
- 5.Liao, P., Chen, T., Chung, P.: A Fast Algorithm for Multilevel Thresholding. Journal of Information Science and Engineering 17, 713–727 (2001)Google Scholar
- 6.Luessi, M., Eichmann, M., Shuster, M., Katsaggelos, A.: New results on efficient optimal multilevel image thresholding. In: Proc. of the IEEE International Conference on Image Processing, Atlanta, USA, pp. 773–776. IEEE Press, Los Alamitos (2006)Google Scholar
- 9.Rueda, L.: A Polynomial-time Algorithm for Optimal Multilevel Thresholding (submitted, 2008), http://www.cs.uwindsor.ca/~lrueda/papers/PolyThresJnl.pdf
- 17.Zhou, Y., Yang, K.: New 2D Adaptive Image Thresholding Method Based on Within and Between Cluster Scatter. In: Proc. of the 27th International Congress on High-Speed Photography and Photonics, SPIE, vol. 6279, p. 62793N (2007)Google Scholar