论文信息 - Image compression based on the neural network art

Image compression based on the neural network art

JPEG-similar algorithms are proposed for compressing video information. These algorithms are based on the ART neural network that realizes the vector quantization operation. The results of modelling the proposed algorithms in the Matlab environment testify to the possibility of using them for image compression. These algorithms are shown to be good enough for images with recurrent segments.

Oleg G. Rudenko | M. S. Snytkin

[1] Philip M. Long,et al. Text compression via alphabet re-representation , 1999, Neural Networks.

[2] B. Yegnanarayana,et al. Artificial Neural Networks , 2004 .

[3] Shawmin Lei,et al. An entropy coding system for digital HDTV applications , 1991, IEEE Trans. Circuits Syst. Video Technol..

[4] Terence D. Sanger,et al. Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[5] O. G. Rudenko,et al. A Generalized Algorithm of Training a Formal Neuron , 2002 .

[6] E. Oja. Simplified neuron model as a principal component analyzer , 1982, Journal of mathematical biology.

[7] Michel Verleysen,et al. Image compression by self-organized Kohonen map , 1998, IEEE Trans. Neural Networks.

[8] N. Ahmed,et al. Discrete Cosine Transform , 1996 .

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

[10] M. V. Velzen,et al. Self-organizing maps , 2007 .

[11] Jürgen Schmidhuber,et al. Sequential neural text compression , 1996, IEEE Trans. Neural Networks.

[12] Stephen Grossberg,et al. Competitive Learning: From Interactive Activation to Adaptive Resonance , 1987, Cogn. Sci..

[13] Robert Hecht-Nielsen,et al. Applications of counterpropagation networks , 1988, Neural Networks.