Title :
A fixed-rate product pyramid vector quantization using a Bayesian model
Author :
Filip, Pavel ; Ruf, Michael J.
Author_Institution :
German Aerosp. Res. Establ., Oberpfaffenhofen, Germany
Abstract :
The performance of the pyramid vector quantizer is analyzed. Some useful formulas for the computation of the pyramid numbers N( L, K) are derived from the generating function. The quantization error, when (L-K)-thresholding is employed is estimated. Extending the Laplacian model, the results can be generalized by applying the Bayesian method. Some simulation results for pyramid vector quantization in DCT (discrete cosine transform) image compression are given
Keywords :
Bayes methods; data compression; discrete cosine transforms; image coding; vector quantisation; (L-K)-thresholding; Bayesian model; DCT image compression; discrete cosine transform; fixed-rate product pyramid vector quantization; quantization error; Bayesian methods; Communications technology; Computational modeling; Data systems; Discrete cosine transforms; Image coding; Laplace equations; Lattices; Performance analysis; Vector quantization;
Conference_Titel :
Global Telecommunications Conference, 1992. Conference Record., GLOBECOM '92. Communication for Global Users., IEEE
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-0608-2
DOI :
10.1109/GLOCOM.1992.276485
