Dynamic Topology Networks for Colour Image Compression
The Self-Organizing Dynamic Graph (SODG) is a novel unsupervised neural network that overcomes some of the limitations of the Kohonen’s Self-Organizing Feature Map (SOFM) by using a dynamic topology among neurons. In this paper an application of the SODG to colour image compression is studied. A Huffman coding and the Lempel-Ziv algorithm are applied to the output of the SODG to provide considerable improvements in compression rates with respect to standard competitive learning. Furthermore, this system is shown to give mean squared errors of the reconstructed images similar to those of competitive learning. Experimental results are presented to illustrate the performance of this system.
KeywordsWeight Vector Image Compression Compression Rate Competitive Learning Huffman Code
Unable to display preview. Download preview PDF.
- 13.Von der Malsburg, C.: Network self-organization. In Zornetzer, S.F., Davis J.L. and Lau C. (eds.): An Introduction to Neural and Electronic Networks. Academic Press, Inc. San Diego, CA (1990), 421–432.Google Scholar
- 14.López-Rubio, E., Muñoz-Pérez, J. and Gómez-Ruiz, J.A.: Self-Organizing Dynamic Graphs. Proceedings of the International Conference on Neural Networks and Applications 2001 (NNA’01), 24–28. N. Mastorakis (Ed.), World Scientific and Engineering Society Press.Google Scholar