Sequências de Baixa Discrepância Aplicadas à Inicialização do Algoritmo Linde-Buzo-Gray

E.A Lima, H. Espírito Santo, F. Madeiro

Abstract


Sistemas de compressão de voz e imagem, esteganografia e reconhecimento de padrões estão entre possíveis aplicações da quantização vetorial (QV),cujo desempenho depende fortemente da qualidade dos dicionários projetados. Uma das técnicas mais amplamente utilizadas para o projeto de dicionários é o algoritmo LBG (Linde-Buzo-Gray). Neste artigo, as sequências de baixa discrepância são apresentadas como uma alternativa de inicialização do algoritmo. O desempenho da técnica apresentada é avaliado no cenário de compressão de imagem baseada em QV.

References


[1] Y.K. Chiang, P. Tsai, Steganography using overlapping codebook partition, Signal Processing, 88, No. 5 (2008), 1203–1215.

[2] I. Dalal, D. Stefan, J. Harwayne-Gidansky, Low discrepancy sequences for Monte Carlo simulations on reconfigurable platforms, In: “Proceedings of the International Conference on Application-Specific Systems, Architectures and

Processors, 2008”, pp. 108–113, 2008.

[3] W.H. Equitz, A new vector quantization clustering algorithm, IEEE Transanctions on Acoustics, Speech and Signal Processing, 37, No. 10 (1989), 1568–1575.

[4] P. Fränti, Genetic algorithm with deterministic crossover for vector quantization, Pattern Recognition Letters, 21, No. 1 (2000), 61–68.

[5] A. Gersho, R.M. Gray, “Vector Quantization and Signal Compression”, Kluwer Academic Publishers, Boston, 1992.

[6] R.M. Gray, Vector quantization, IEEE ASSP Magazine, (1984), 4–29.

[7] J.H. Halton, On the efficiency of certain quasi-random sequences of points in evaluating multidimensional integrals, Numer. Math., (1960), 84–90.

[8] Y. Hu, High–capacity image hiding scheme based on vector quantization, Pattern Recognition, 3, (2006), 1715–1724.

[9] I. Katsavounidis, C.-C.J. Kuo, Z. Zhang, A new initialization technique for geneneralized lloyd iteration, IEEE Signal Processing Letters, 1, No. 10, (1994), 144–146.

[10] A. Kjoelen, S.E. Umbaugh, M. Zuke, Compression of skin tumor images – wavelet/vector quantization methods for reducing the time, cost and bandwidth of storing and transmitting data, IEEE Engineering in Medicine and Biology, (1998), 73–80.

[11] N.H. Kuipers, “Uniform Distribution of Sequences”, AddisonWesley Publishing Company, Inc., New York, 1992.

[12] Y. Linde, A. Buzo, R.M. Gray, An algorithm for vector quantizer design, IEEE Transactions on Communications, 28, No. 1 (1980), 84–95.

[13] W.T.A. Lopes, F. Madeiro, B.G. Aguiar Neto, M.S. Alencar, Combining modulation diversity and index assignment to improve image VQ for a Rayleigh fading channel, Learning and Nonlinear Models – Revista da Sociedade Brasileira de Redes Neurais, 1, No. 3 (2004), 160–171.

[14] F. Madeiro, J.M. Fechine, W.T.A. Lopes, B.G. Aguiar Neto, M.S. Alencar, Identificação Vocal por Frequência Fundamental, QV e HMMS, In: “Em- TOM-Ação: A Prosódia em Perspectiva”, (M.A.M. Aguiar and F. Madeiro, eds.), chapter 4, pp. 91–120. Editora Universitária da UFPE, Recife, PE, first edition, 2007.

[15] K.K. Paliwal, B.S. Atal, Efficient vector quantization of LPC parameters at 24 bits/frame, IEEE Transactions on Speech and Audio Processing, 1, No. 1 (1993), 3–14.

[16] O. Teytaud, When does quasi-random work?, In: “Proceedings of the 10th International Conference on Parallel Problem Solving from Nature: PPSN X”, pp. 325–336, 2008.

[17] Van der Corput, Zahlentheoretische abschaätzungen, Math. Ann., (1921), 53–79.

[18] P. Yahampath, P. Rondeau, Multiple-description predictive-vector quantization with applications to low bit-rate speech coding over networks, IEEE Transactions on Audio, Speech, and Language Processing, 15, No. 3 (2007), 749–755.




DOI: https://doi.org/10.5540/tema.2010.011.03.0217

Article Metrics

Metrics Loading ...

Metrics powered by PLOS ALM

Refbacks

  • There are currently no refbacks.


Trends in Computational and Applied Mathematics

A publication of the Brazilian Society of Applied and Computational Mathematics (SBMAC)

 

Indexed in:

                       

         

 

Desenvolvido por:

Logomarca da Lepidus Tecnologia