Abstract
Neural-like approach to the thinning of binary patterns based on local thickness estimate is proposed. Thinning is considered to be a self-organizing process where a binary neuron — contour pixel value is given by the hard limiter transfer function. Input to this function is given by the pattern local thickness estimate from flexible (both in size and shape) neighborhood together with neuron threshold which ensures the preservation of the 4-connectedness and genuine end point. The final skeletons obtained within sequential, hybrid and parallel mode of operation maintain perfectly connectedness and topology. Due to the presence of the more global and structural information about pattern (involved in the local thickness estimate) the enhanced preservation of geometric properties is observed.
Preview
Unable to display preview. Download preview PDF.
References
Ahmed, P.: A neural network based dedicated thinning method. Pattern Recognition Letters 16, (1995) 1267–1275.
Bruske, J., Sommer G.: Dynamic cell structure learns perfectly topology preserving map. Neural Computation 7, (1995) 845–865.
Farkaš, I., Chudý, L.: Application of a growing self-organising map to thinning of binary patterns with noise. Accepted in Workshop on Self-organising Maps, Helsinki, June 1997
Hilditch, C.J.: Linear skeletons from square cupboards. In: B.Meltzer and D.Michie, Eds., Machine Intell., New York: Amer. Elsevier, vol. 4, (1969) 403–420
Kohonen, T.: Self-organizing maps. Springer Verlag (1995).
Lam, L., Lee, S.W., Suen, C.Y.: Thinning Methodologies — A Comprehensive Survey. IEEE Trans. Pattern Anal. Mach. Intell. 14, (1992) 869–885.
Matsumoto, T., Chua, L.O., and Yokohama, T.: Image Thinning witha Cellular Neural Network. IEEE Trans. Circ. Sys. 37, (1990) 638–640.
Rosenblatt, F. The Perceptron: A probabilistic model for information storage and organization in the brain. Psychological Review 65, (1958) 386–408
Rutowitz, D.: Pattern recognition. J.Roy.Stat.Soc., 129, Series A, (1966) 504–530.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chudý, L., Chudý, V. (1997). Neural-like thinning processing. In: Sommer, G., Daniilidis, K., Pauli, J. (eds) Computer Analysis of Images and Patterns. CAIP 1997. Lecture Notes in Computer Science, vol 1296. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63460-6_161
Download citation
DOI: https://doi.org/10.1007/3-540-63460-6_161
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63460-7
Online ISBN: 978-3-540-69556-1
eBook Packages: Springer Book Archive