A neural network architecture with self-organization in phase and actual space is proposed and discussed. Special type of differential local interconnections simulating diffusion, dispersion, and convection were investigated. It is shown that these interconnections are responsible for biological pattern formation in a homogeneous neural structure. The model suggests a phenomenological explanation of the mechanisms of edge detection in vision process.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
Amari S (1983) Field theory of self-organizing neural nets. IEEE Trans Syst Man Cybern 13:741–748
Appel P (1953) Traite de mecanique rationelle, Part II Sect. XXV. Gauthier-Villars, Paris
Hurlbert A, Poggio T (1989) Making machines (and art int). See Art. Int. Debate, pp 213–239
Sakaguchi Y (1990) Topographic organization of nerve field with teacher signal. Neural Network 3:411–421
Whitham G (1972) Linear and nonlinear waves. Wiley, New York, pp 19–25, 460–466
Zak M (1989a) Terminal attractors in neural networks. Neural Networks 2:259–274
Zak M (1989b) The least constraint principle for learning in neurodynamics. Phys Lett A 135:25–28
About this article
Cite this article
Zak, M. Neurodynamics with spatial self-organizations. Biol. Cybern. 65, 121–127 (1991). https://doi.org/10.1007/BF00202387
- Neural Network
- Edge Detection
- Pattern Formation
- Network Architecture