Image Compression using an Enhanced Self Organizing Map Algorithm with Vigilance Parameter

In this paper, a new approach for image compression is presented. The enhanced SOM algorithm applies a vigilance parameter in order to test if the maximum value of all activation functions in each training step exceeds the minimum threshold. If vigilance test is not passed, a new cluster will be added to the network; else the winner cluster will be updated. Therefore, the network could be extendable due to pattern distribution. The proposed approach is compared with ART1. The performance of enhanced SOM algorithm does not depend on the input presentation order. As observed through simulations, the new algorithm with vigilance parameter reduces the computational complexity, and presents better quality compared with Kohonen's SOM.

[1]  Teuvo Kohonen,et al.  Self-Organizing Maps, Second Edition , 1997, Springer Series in Information Sciences.

[2]  Yan Shi,et al.  A kind of improved method of fuzzy clustering , 2005, 2005 International Conference on Machine Learning and Cybernetics.

[3]  Vladimir Estivill-Castro,et al.  Why so many clustering algorithms: a position paper , 2002, SKDD.

[4]  Jzau-Sheng Lin,et al.  The application of competitive Hopfield neural network to medical image segmentation , 1996, IEEE Trans. Medical Imaging.

[5]  Estivill-CastroVladimir.  Why so many clustering algorithms , 2002 .

[6]  Zhi-Hua Zhou,et al.  SOM Ensemble-Based Image Segmentation , 2004, Neural Processing Letters.

[7]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[8]  阿部 重夫.  Pattern classification : neuro-fuzzy methods and their comparison , 2001 .

[9]  Michael Chester,et al.  Neural networks - a tutorial , 1993 .

[10]  Roberto Tagliaferri,et al.  Automated labeling for unsupervised neural networks: a hierarchical approach , 1999, IEEE Trans. Neural Networks.

[11]  Melody Y. Kiang,et al.  The Effect of Sample Size on the Extended Self-Organizing Map Network for Market Segmentation , 2005, Proceedings of the 38th Annual Hawaii International Conference on System Sciences.

[12]  Ernest L. Hall,et al.  Computer Image Processing and Recognition , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Stephen Grossberg,et al.  The ART of Adaptive Pattern Recognition Self-organizing by a Neu Network , 1988 .

[14]  Luigi Cinque,et al.  A modified fuzzy ART for image segmentation , 2001, Proceedings 11th International Conference on Image Analysis and Processing.

[15]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .