The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Este artigo apresenta um novo algoritmo de aprendizagem competitiva para o projeto de quantizadores vetoriais de taxa variável (VQs). O algoritmo, denominado algoritmo de aprendizagem competitiva de taxa variável (VRCL), projeta um VQ com distorção média mínima sujeita a uma restrição de taxa. O VRCL realiza o treinamento do vetor de pesos no domínio wavelet para que o tempo de treinamento necessário seja curto. Além disso, o algoritmo apresenta um melhor desempenho de distorção de taxa do que outros algoritmos de design VQ existentes e algoritmos de aprendizagem competitivos. O algoritmo de aprendizagem também é mais insensível à seleção de palavras-código iniciais em comparação com algoritmos de design existentes. Portanto, o algoritmo VRCL pode ser uma alternativa eficaz aos algoritmos de projeto VQ de taxa variável existentes para aplicações de compressão de sinal.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Wen-Jyi HWANG, Maw-Rong LEOU, Shih-Chiang LIAO, Chienmin OU, "A Novel Competitive Learning Technique for the Design of Variable-Rate Vector Quantizers with Reproduction Vector Training in the Wavelet Domain" in IEICE TRANSACTIONS on Information,
vol. E83-D, no. 9, pp. 1781-1789, September 2000, doi: .
Abstract: This paper presents a novel competitive learning algorithm for the design of variable-rate vector quantizers (VQs). The algorithm, termed variable-rate competitive learning (VRCL) algorithm, designs a VQ having minimum average distortion subject to a rate constraint. The VRCL performs the weight vector training in the wavelet domain so that required training time is short. In addition, the algorithm enjoys a better rate-distortion performance than that of other existing VQ design algorithms and competitive learning algorithms. The learning algorithm is also more insensitive to the selection of initial codewords as compared with existing design algorithms. Therefore, the VRCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.
URL: https://global.ieice.org/en_transactions/information/10.1587/e83-d_9_1781/_p
Copiar
@ARTICLE{e83-d_9_1781,
author={Wen-Jyi HWANG, Maw-Rong LEOU, Shih-Chiang LIAO, Chienmin OU, },
journal={IEICE TRANSACTIONS on Information},
title={A Novel Competitive Learning Technique for the Design of Variable-Rate Vector Quantizers with Reproduction Vector Training in the Wavelet Domain},
year={2000},
volume={E83-D},
number={9},
pages={1781-1789},
abstract={This paper presents a novel competitive learning algorithm for the design of variable-rate vector quantizers (VQs). The algorithm, termed variable-rate competitive learning (VRCL) algorithm, designs a VQ having minimum average distortion subject to a rate constraint. The VRCL performs the weight vector training in the wavelet domain so that required training time is short. In addition, the algorithm enjoys a better rate-distortion performance than that of other existing VQ design algorithms and competitive learning algorithms. The learning algorithm is also more insensitive to the selection of initial codewords as compared with existing design algorithms. Therefore, the VRCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.},
keywords={},
doi={},
ISSN={},
month={September},}
Copiar
TY - JOUR
TI - A Novel Competitive Learning Technique for the Design of Variable-Rate Vector Quantizers with Reproduction Vector Training in the Wavelet Domain
T2 - IEICE TRANSACTIONS on Information
SP - 1781
EP - 1789
AU - Wen-Jyi HWANG
AU - Maw-Rong LEOU
AU - Shih-Chiang LIAO
AU - Chienmin OU
PY - 2000
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E83-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2000
AB - This paper presents a novel competitive learning algorithm for the design of variable-rate vector quantizers (VQs). The algorithm, termed variable-rate competitive learning (VRCL) algorithm, designs a VQ having minimum average distortion subject to a rate constraint. The VRCL performs the weight vector training in the wavelet domain so that required training time is short. In addition, the algorithm enjoys a better rate-distortion performance than that of other existing VQ design algorithms and competitive learning algorithms. The learning algorithm is also more insensitive to the selection of initial codewords as compared with existing design algorithms. Therefore, the VRCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.
ER -