%0 Conference Proceedings
%@isbn 978-85-7669-270-6
%A Rila, Luciano,
%A Alcaim, Abraham,
%@affiliation IBM Brasil – Centro Científico Rio
%@affiliation IBM Brasil – Centro Científico Rio
%T Codificação de imagens em sub-bandas com quantização vetorial inter-bandas particionada
%B Simpósio Brasileiro de Computação Gráfica e Processamento de Imagens, 5 (SIBGRAPI)
%D 1992
%E Câmara, Gilberto,
%E Gomes, Jonas de Miranda,
%V 1
%S Anais
%8 10 - 12 nov. 1992
%J Porto Alegre
%I Sociedade Brasileira de Computação
%C Águas de Lindóia
%K codificação de imagens, sub-bandas, quantização vetorial, inter-bandas, processamento de imagem.
%X Subband coding has proven to be an efficient method for coding images at low bit rates. In this paper we consider an algorithm to compress still images which is based on subband coding and inter-band vector quantization. The image spectrum is decomposed into 16 uniform bands using 2-D separable quadrature mirror filters. These 16 bands are used to create 16-dimensional vectors, which are coded using vector quantization. Our aprouch is to split these vectors in order to reduce complexity and exploit spectral properties of the image data. We examine the effects of the bit allocation at rates of 0.50 bit/pixel and 0.63 bit/pixel. Simulation results show that the performance is comparable to the basic coding structure, while complexity is significantly reduced.
%P 39-45
%@language pt
%9 Processamento de Imagens
%3 5 Codificacao de imagens em sub_bandas com quantizacao vetorial inter_bandas particionada.pdf