%0 Conference Proceedings
%4 sid.inpe.br/banon/2002/11.06.11.36
%2 sid.inpe.br/banon/2002/11.06.11.36.23
%@doi 10.1109/SIBGRA.2000.883912
%T Fractal color image compression
%D 2000
%A Zhaohui, Li,
%A Liang, Zhao,
%A Soma, Nei Y.,
%E Carvalho, Paulo Cezar Pinto,
%E Walter, Marcelo,
%B Brazilian Symposium on Computer Graphics and Image Processing, 13 (SIBGRAPI)
%C Gramado, RS, Brazil
%8 17-20 Oct. 2000
%I IEEE Computer Society
%J Los Alamitos
%P 185-192
%S Proceedings
%1 SBC - Brazilian Computer Society
%K data compression fractal color image compression Fractal Hierarchical Color Block Coding FHCBC three-color planes one-component image homogeneous blocks pixel trichromatic coefficients threshold value mean value fractal coding decoding process fractal decoding algorithm mean values trichromatic coefficient ratios three-component Separated Fractal Coding performance gains running time compression ratio reconstruction quality.
%X The paper presents a new fractal color image compression method, called Fractal Hierarchical Color Block Coding (FHCBC), which transforms the three-color planes of a color image into a one-component image by extracting correlation among them. It hierarchically divides the three-color planes into homogeneous blocks, and for each such block, the variance of pixels' trichromatic coefficients is within a certain threshold value. Then, each block is represented by its mean value of the pixels' trichromatic coefficient ratios, and just a one-component image is composed and compressed by fractal coding. In the decoding process, the reconstruction is carried out by a fractal decoding algorithm and the associated mean values of trichromatic coefficient ratios. In comparison with the well known three-component Separated Fractal Coding (SFC), the suggested method generates performance gains in running time, high compression ratio and equal level of reconstruction quality.
%3 185-192.pdf
%O The conference was held in Gramado, RS, Brazil, from October 17 to 20.