Summary
International Symposium on Nonlinear Theory and its Applications
2010
Session Number:C1L-B
Session:
Number:C1L-B1
Associative Dynamics of Color Images in a Chaotic Neural Network
Makito Oku, Kazuyuki Aihara,
pp.465-468
Publication Date:2010/9/5
Online ISSN:2188-5079
Summary:
Recently, we have succeeded in numerically simulating a large-scale chaotic neural network with 1 million units. This success will open new possibilities for applications of chaotic neural networks from the viewpoints of academic interest as well as new media art. In this paper, we report a way to deal with color images by using a large-scale chaotic neural network. In the proposed method, color images are converted to binary sequences, modified slightly by inverting some bits, and stored in the network. The results of numerical simulations show that chaotic alternations among stored patterns as well as their reverse patterns can be observed within a certain range of parameters. We also compare four di?erent coding schemes of color information, which changes the appearance of chaotic dynamics.