Minimum loss of information and image segmentation
A new measure of the information loss in image segmentation is derived from a set of natural properties. A similar quantity can be used in the quantization of a continuous real random n-vector. A new method for thresholding the grey-level histogram of a picture is then introduced. The method is based on the natural requirement of minimum information loss.
Keywordsquantization image segmentation information loss entropy
Unable to display preview. Download preview PDF.
- Abutaleb, A.S. (1989) “Automatic Thresholding Grey-Level Pictures Using Two-Dimensional Entropy”, Comput. Vision, Graphics Image Process., 42, pp. 22–31.Google Scholar
- Aczél, J. (1978) “A mixed Theory of Information — II: Additive Inset Entropies (of Randomized systems of Events) with Measurable Sum Property”, Utilitas Mathematica, Vol. 13, pp. 49–54.Google Scholar
- Aczél, J. and Daroczy, Z. (1975) “On measures of Information and their Characterizations”, Academic Press, New York-San Francisco-London.Google Scholar
- Caselli, R. and Forte, B. (1988) “Thresholding Grey-Level Histograms by Minimum Information Loss”, SASIAM internal report.Google Scholar
- Forte, B. and Kolbas, V. “Some Experimental Results in Image Segmentation by Minimum Loss of Information”, manuscript.Google Scholar
- Forte, B., Ng, C.T. and Lo Schiavo, M. (1984) “Additive and Subadditive Entropies for Discrete Random Vectors”, Journal of Comb. Info. and Systems Sc., Vol. 9, No. 4, 207–216.Google Scholar
- Forte, B. and Sahoo, P.K., “Minimal Loss of Information and Optimal Thresholds for Digital Images”, manuscript.Google Scholar
- Sahoo, P.K., Soltani, S., Wong, A.K.C. and Chen, Y.C. (1988) “A Survey of Thresholding Techniques”, Computer Vision, Graphics and Image Process., 41, 233–260.Google Scholar
- Wong, A.K.C. and Sahoo, P.K. (1989) “A Grey-Level Threshold Selection Method Based on Maximum Entropy Principle”, IEEE Trans. on Systems, Man and Cybernetics, Vol. 19, No. 4, July/August 866–871.Google Scholar