Title :
Image compression via optimal vector quantization: a comparison between SOM, LBG and k-means algorithms
Author :
Corral, Juan A. ; Guerrero, Miguel ; Zufiria, Pedro J.
Author_Institution :
ETSI Telecomunicacion, Univ. Politecnica de Madrid, Spain
fDate :
27 Jun- 2 Jul 1994
Abstract :
An application of optimal vector quantization on image compression is studied. A neural network structure for obtaining the optimal codebook for the vector quantizer (VQ) is employed. This structure is based on Kohonen´s self-organizing map (SOM), whose learning algorithm provides an optimal codebook for a training sequence. It is demonstrated that the SOM complies, in general, with the Max-Lloyd´s conditions for optimal VQ. In this line, it is shown that the obtained codebook minimizes the averaged distortion over the training sequence, provided that the certain regularity conditions are satisfied. Finally, SOM-VQ convergence properties and squared-mean-error results are compared with LBG as well as k-means algorithms
Keywords :
image coding; image processing; learning (artificial intelligence); optimisation; self-organising feature maps; vector quantisation; Kohonen self-organizing map; LBG algorithm; Max-Lloyd conditions; codebook; convergence; k-means algorithm; learning algorithm; optimal vector quantization; squared-mean-error; training sequence; Costs; Data compression; Image coding; Image converters; Image quality; Image storage; Neural networks; Neurons; Telecommunications; Vector quantization;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374873
