(14-15(01)) * << * >> * Russian * English * Content * All Issues

Neural networks based on multi-valued neural elements: learning, image processing and recognition
N.N. Aizenberg, I.N. Aizenberg, G.A. Krivosheev

 PDF, 2302 kB

Pages: 179-186.

Full text of article: Russian language.

Abstract:
This work considers the development of the concept of multi-valued neural elements. First of all, a significantly enhanced learning algorithm is proposed, which allows to accelerate the learning process by 20-30 times and to achieve training convergence even for the functions for which the successful learning used to be considered impossible. We consider the networks of multi-valued neural elements, focused primarily on solving the issues of processing (cell networks) and recognition (networks with random connections) of images. We show the ways to solve the tasks of detecting edges, analyzing textures, and developing associative memory of half-tone images using the above networks.

Citation:
Aizenberg NN, Aizenberg IN, Krivosheev GA. Neural networks based on multi-valued neural elements: learning, image processing and recognition. Computer Optics 1995; 14-15(1): 179-
186.

References:
  1. Aizenberg NN, Ivaskiv YL. Multiple-valued threshold logic, Kiev: Naukova Dumka; 1977
  2. Aizenberg NN, Aizenberg IN. CNN Based on Multi-Valued Neuron as a Model of Associative Memory for Gray-Scale Images. Proc. of the 2-d International IEEE Workshop on Cellular Neural Networks and their Applications; Germany; Munich: 1992; IEEE 92TH0498-6; ISBN 0-7803-875-1; 36-41.
  3. Aizenberg NN, Aizenberg IN. Fast Convergenced Learning Algorithms for Multi-Level and Universal Binary Neurons and Solving of the some Image Processing Problems, Lecture Notes in Computer Science; Ed. Mira J, Cabestany J, Prieto A. Shpringer- Verlag, Berlin-Heidelberg: 1993, 686: 230-236.
  4. Aizenberg NN, Aizenberg IN. CNN-like Networks Based on Multi-Valued and Universal Binary Neurons: Learning and Application to Image Processing", Proc. of the Third IEEE International Workshop on Cellular Neural Networks and their Applications; Italy; Roma: 1994; IEEE 94TH0693-2; 153-158.
  5. Chua LO, Yang L. Cellular neural networks: Theory; IEEE Trans.Circuits Syst.; 1988; 35: 1257-1290.
  6. Aizenberg IN. Universal Logical Element over the Field of the Complex Numbers; Kibernetika (Cybernetics); 1991; 3: 116-121.
  7. Kohonen T. Content-Addressable Memories; Springer-Verlag; Berlin-Heidelberg-New-York: 1980.
  8. Ramaher U, Raab W, Anlauf J, Hachmann U, Beichter J, Bruls N, Weseling M, Sichender E, Manner R, Glass J, Wurz A. Multiprocessor and Memory Architecture of the Neurocomputer Synapse-1; Proceedings of the 3-d International Conference on Microelectronics for Neural Networks; UK; Edinburgh: 1993; 227-231 .

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20