TY - GEN
T1 - Application of neural `gas' model in image compression
AU - Zhang, Bai ling
AU - Fu, Min yue
AU - Yan, Hong
PY - 1998
Y1 - 1998
N2 - Topologically ordered vector quantization has attracted attention in recent years. In this paper, we propose to apply a topology representing learning algorithm - neural `gas' model for obtaining topology ordered codebook for the vector quantization (VQ) and exploit it on image compression. Compared with the well-known Kohonen's self-organizing map (SOM), neural `gas' model has several advantages, including faster convergence and higher signal-to-noise ratio in reconstruction. We illustrate some experimental results and discuss several relevant research issues.
AB - Topologically ordered vector quantization has attracted attention in recent years. In this paper, we propose to apply a topology representing learning algorithm - neural `gas' model for obtaining topology ordered codebook for the vector quantization (VQ) and exploit it on image compression. Compared with the well-known Kohonen's self-organizing map (SOM), neural `gas' model has several advantages, including faster convergence and higher signal-to-noise ratio in reconstruction. We illustrate some experimental results and discuss several relevant research issues.
UR - https://www.scopus.com/pages/publications/0031634789
UR - https://www.scopus.com/pages/publications/0031634789#tab=impact
M3 - RGC 32 - Refereed conference paper (with host publication)
VL - 2
SP - 918
EP - 920
BT - IEEE International Conference on Neural Networks - Conference Proceedings
PB - IEEE
T2 - 1998 IEEE International Joint Conference on Neural Networks (IJCNN 1998)
Y2 - 4 May 1998 through 9 May 1998
ER -