Neural Processing Letters

, Volume 34, Issue 1, pp 39–58 | Cite as

A Skeletonizing Reconfigurable Self-Organizing Model: Validation Through Text Recognition



Self Organizing Maps are able to develop topology preserving classifiers. In this work we propose a Reconfigurable Self Organizing Model, which combines this property with others related with the generation of sub-graphs of the Delaunay-triangulation, the possibility of generating elastic approximations and the capacity to reconfigure the models topological structure in a data driven way. These properties allow us to apply the model to the extraction of linear structures from one-dimensional curves and from two-dimensional figures (which can be dense or not). Skeletonization and recognition of machine printed text and handwritten numerals serve as a validation domain.


Neural networks Self-organizing map Machine printed text Handwritten text recognition Skeletonization 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Kohonen T (1982) Self-organized formation of topologically correct feature maps. Biol Cybern 43(2): 59–69MATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    Kohonen T (1989) Self-organization and associative memory. Springer Verlag, BerlinGoogle Scholar
  3. 3.
    Kohonen T (2001) Self-organizing maps, Vol 30 of Springer series in information sciences. 3. Springer Verlag, Berlin-HeidelbergGoogle Scholar
  4. 4.
    Angéniol B, DeLaCroix Vaubois G, Le Texier J-I (1988) Self-organizing feature maps and the traveling salesman problem. Neural Netw 1: 289–293CrossRefGoogle Scholar
  5. 5.
    Durbin R, Willshaw D (1987) An analogue approach to the travelling salesman problem using an elastic net method. Nature 326: 689–691CrossRefGoogle Scholar
  6. 6.
    Alonso-Weber JM, Sanchis de Miguel A (2006) Off-line cursive handwritten text outline drawing with self-organizing maps. In: 2006 International conference on engineering and mathematics (ENMA 2006), 10–11 July 2006, Bilbao, SpainGoogle Scholar
  7. 7.
    Plamondon R, Srihari SN (2000) On-line and off-line handwriting recognition: a comprehensive survey. IEEE Trans Pattern Anal Mach Intell 22(1): 63–84CrossRefGoogle Scholar
  8. 8.
    Steinherz T, Rivlin E, Intrator N (1999) Off-line cursive script word recognition—a survey. Int J Doc Anal Recognit. 2(2–3): 90–110Google Scholar
  9. 9.
    Blum H (1967) A transformation for extracting new descriptors of shape. In: Wathen-Dunn W (edS) Symposium on models for the perception of speech and visual form. MIT Press. pp 362–380Google Scholar
  10. 10.
    Singh R, Cherkassky V, Papanikolopoulos NP (2000) Self-organizing maps for the skeletonization of sparse shapes. IEEE Trans Neural Netw 11(1): 241–248CrossRefGoogle Scholar
  11. 11.
    Ogniewicz R, Kübler O (1995) Hierarchical voronoi skeletons. Pattern Recognit 28(3): 343–359CrossRefGoogle Scholar
  12. 12.
    Hastie T, Stuetzle W (1989) Principal curves. J Am Stat Assoc 84(406): 502–516MATHCrossRefMathSciNetGoogle Scholar
  13. 13.
    Kégl B (1999) Principal curves: learning, design, and applications. PhD Thesis, Concordia University, Montreal, CanadaGoogle Scholar
  14. 14.
    Kégl B, Krzyzak A (2002) Piecewise linear skeletonization using principal curves. IEEE Trans Pattern Anal Mach Intell 24(1): 59–74CrossRefGoogle Scholar
  15. 15.
    Yin H (2007) Nonlinear dimensionality reduction and data visualization: a review. Int J Autom Comput 4(3): 294–303CrossRefGoogle Scholar
  16. 16.
    Hilditch C (1969) Linear skeletons from square cupboards. In: Meltzer B, Mitchie D (eds) Machine Intelligence. pp 403–420Google Scholar
  17. 17.
    Pavlidis T (1980) A Thinning Algorithm for discrete binary images. Comput Graph Image Process 13(2): 142–157CrossRefMathSciNetGoogle Scholar
  18. 18.
    Naccache NJ, Shingal R (1984) SPTA: a proposed algorithm for thinning binary patterns. IEEE Trans Syst Man Cybern 14(3): 409–418Google Scholar
  19. 19.
    Suzuki S, Abe K (1986) Sequential thinning of binary pictures using distance transformation. In: Proceedings of the 8th international conference on pattern recognition, pp 289–292Google Scholar
  20. 20.
    Klette G (2007) Topologic, geometric, or graph-theoretic properties of skeletal curves. PhD Thesis, Faculty of Mathematics and natural Sciences, Univ. GroningenGoogle Scholar
  21. 21.
    Chen YS, Yu Yu T (1996) Thinning approaches for noisy digital patterns. Pattern Recognit 29(11): 1847–1862CrossRefGoogle Scholar
  22. 22.
    Martinetz T, Schulten K (1991) A neural gas network learns topologies. In: Kohonen T et al (eds) IEEE international conference on artificial neural networks, Espoo, Finland, vol 1. Elsevier, pp 397–407Google Scholar
  23. 23.
    Datta A, Paruri SK, Chaudhuri BB (1996) Skeletal Shape Extraction from Dot Patterns by Self-Organization. In: Proceedings of the 13th internacional conference on pattern recognition 4:80-84Google Scholar
  24. 24.
    Singh R, Cherkassky V, Papanikolopoulos NP (1997) Determining the skeletal description of sparse shapes. Proc IEEE Int Symp Comput Intell Robot Autom 368–373Google Scholar
  25. 25.
    Singh R, Papanikolopoulos NP, Cherkassky V (1998) Object skeletons from sparse shapes in industrial image settings. In: Proc IEEE Int Conf Robotics Autom 4:3388–3393Google Scholar
  26. 26.
    Singh R, Wade MC, Papanikolopoulos NP (1998) Letter-level shape description by skeletonization in faded documents. In: Proceedings of the Fourth IEEE Workshop on Applications of Computer Vision. IEEE Comput. Soc. Press, pp 121–126Google Scholar
  27. 27.
    Kangas JA, Kohonen T, Laaksonen T (1990) Variants of self-organizing maps. IEEE Trans Neural Netw 1: 93–99CrossRefGoogle Scholar
  28. 28.
    Kégl B, Krzyzak A, Linder T, Zeger K (2000) Learning and design of principal curves. IEEE Trans Pattern Anal Mach Intell 22(3): 281–297CrossRefGoogle Scholar
  29. 29.
    Martinetz TM, Schulten KJ (1994) Topology representing networks. Neural Netw 7(3): 507–522CrossRefGoogle Scholar
  30. 30.
    Fritzke B (1995) A growing neural gas network learns topologies. Adv Neural Inf Process Syst 7: 625–632Google Scholar
  31. 31.
    Fritzke B (1996) Growing self-organizing networks—why?. In: Verleysen M (eds) ESANN’96: European symposium on artificial neural networks. D-Facto Publishers, Brussels, pp 61–72Google Scholar
  32. 32.
    Fritzke B (1994) Growing cell structures—a self-organizing network for unsupervised and supervised learning. Neural Netw 7(9): 1441–1460CrossRefGoogle Scholar
  33. 33.
    LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324. Google Scholar
  34. 34.
    Bernard TM, Manzanera A (1999) Improved low complexity fully parallel thinning algorithm. In: Proceedings of the 10th international conference on image analysis and processing, IEEE computer societyGoogle Scholar
  35. 35.
    Camastra F, Vinciarelli A (2001) Cursive character recognition by learning vector quantization. Pattern Recognit Lett 22(6-7): 625–629MATHCrossRefGoogle Scholar
  36. 36.
    Camastra F, Spinetti M, Vinciarelli A (2006) Offline cursive character challenge: a new benchmark for machine learning and pattern recognition algorithms. In: 18th International conference on pattern recognition (ICPR 2006), 20–24 August 2006, Hong Kong, China. IEEE Computer Society 2006, pp 913–916Google Scholar
  37. 37.
    Maiorana F (2008) Time and space complexity improvements for a grid implementation of a kohonen-like classification algorithms on sparse data-sets. Int J Math Comput Simul 2(4): 328–337Google Scholar
  38. 38.
    Ancona F, Ridella S, Rovetta S, Zunino R (1997) On the importance of sorting in ‘Neural Gas’ training of vector quantizers. Proceedings of the IEEE International Conference of Neural Networks, pp 1804–1807Google Scholar

Public Domain Software:

  1. 39.
    GNU Ocrad (Optical Character Recognition Program)
  2. 40.

Copyright information

© Springer Science+Business Media, LLC. 2011

Authors and Affiliations

  1. 1.Department of Informatics, Polytechnic SchoolUniversidad Carlos III de MadridLeganés, MadridSpain

Personalised recommendations