Incremental Self-Organizing Map (iSOM) in Categorization of Visual Objects

  • Andrew P. Papliński
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7664)


We present a modification of the well-known Self-Organizing Map (SOM) in which we incrementally allocate the neuronal nodes to progressively added new stimuli. Our incremental SOM (iSOM) aims at the situation when a stimulus, or percept, is represented by a number of neuronal nodes a typical case in biological situation when the redundancy of representation of data is important. The iSOM is applied to categorization of visual objects using the recently introduced feature vector based on the angular integral of the Radon transform [10].


Self-organizing maps Incremental learning Radon transform 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Alahakoon, D., Halgamuge, S.K., Srinivasan, B.: Dynamic Self-organizing Maps with Controlled Growth for Knowledge Discovery. IEEE Trans. Neural Networks 11(3), 601–614 (2000)CrossRefGoogle Scholar
  2. 2.
    Blackmore, J., Miikkulainen, R.: Incremental Grid Growing: Encoding High-Dimensional Structure into a Two-Dimensional Feature Map. In: Proc. IEEE Int. Conf. Neural Networks, pp. 450–455 (1993)Google Scholar
  3. 3.
    Bosch, A., Zisserman, A., Munoz, X.: Image Classification Using Random Forests and Ferns. In: 11th Int. Conf. Computer Vision, vol. 21(4), pp. 1–8 (2007)Google Scholar
  4. 4.
    Fritzke, B.: Growing Cell Structures — a Self-Organizing Network for Unsupervised and Supervised Learning. Neural Networks 7(9), 1441–1460 (1994)CrossRefGoogle Scholar
  5. 5.
    Fritzke, B.: A Growing Natural Gas Network Learns Topologies. In: Advances in Neural Information Processing, vol. 7, pp. 625–632. MIT Press (1995)Google Scholar
  6. 6.
    Jantvik, T., Gustafsson, L., Papliński, A.P.: A Self-Organized Artificial Neural Network Architecture for Sensory Integration with Applications to Letter–Phoneme Integration. Neural Computation 23, 2101–2139 (2011)CrossRefGoogle Scholar
  7. 7.
    Kohonen, T.: Self-Organising Maps, 3rd edn. Springer, Berlin (2001)CrossRefGoogle Scholar
  8. 8.
    Lee, J.A., Verleysen, M.: Nonlinear Dimensionality Reduction. Springer (2007)Google Scholar
  9. 9.
    Lowe, D.G.: Distinctive Image Features From Scale-Invariant Keypoints. Int. J. Comp. Vision 60(2), 91–110 (2004)CrossRefGoogle Scholar
  10. 10.
    Papliński, A.P.: Rotation Invariant Categorization of Colour Images Using Radon Transform. In: Proc. WCCI–IJCNN, pp. 1408–1413. IEEE (2012)Google Scholar
  11. 11.
    Papliński, A.P., Gustafsson, L., Mount, W.M.: A Model of Binding Concepts to Spoken Names. Aust. Journal of Intelligent Information Processing Systems 11(2), 1–5 (2010)Google Scholar
  12. 12.
    Papliński, A.P., Gustafsson, L., Mount, W.M.: A Recurrent Multimodal Network for Binding Written Words and Sensory-Based Semantics into Concepts. In: Lu, B.-L., Zhang, L., Kwok, J. (eds.) ICONIP 2011, Part I. LNCS, vol. 7062, pp. 413–422. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  13. 13.
  14. 14.
    Sammon, J.W.: A Nonlinear Mapping for Data Structure Analysis. IEEE Trans. Computers 18(5), 401–409 (1969)CrossRefGoogle Scholar
  15. 15.
    Serre, T., Wolf, L., Poggio, T.: Object Recognition with Features Inspired by Visual Cortex. In: IEEE Comp. Soc. Conf. Comp. Vision and Patt. Recognition, vol. 2, pp. 994–1000 (2005)Google Scholar
  16. 16.
    Shah-Hosseini, H., Safabakhsh, R.: TASOM: A New Time Adaptive Self-Organizing Map. IEEE Trans. Syst., Man, Cyber. B 33(2), 271–282 (2003)CrossRefGoogle Scholar
  17. 17.
    Shah-Hosseini, H., Safabakhsh, R.: Binary Tree Time Adaptive Self-Organizing Map. Neurocomputing 74, 1823–1839 (2011)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Andrew P. Papliński
    • 1
  1. 1.Monash UniversityAustralia

Personalised recommendations