Title :
Symmetry-adapted wavelet transform image compression
Author :
Levy, I. ; Wilson, Richard
Author_Institution :
Dept. of Comput. Sci., Warwick Univ., Coventry, UK
Abstract :
A novel wavelet decomposition, based on a set of six oriented bands in each scale, is presented and its use in image data compression described. The new decomposition uses complex wavelets derived from the conventional separable dyadic transform. The new representation is used in a vector quantizer (VQ), which exploits the symmetry relations which exist between the bands in different scales and orientations. The resulting coder performance is competitive with schemes such as embedded zerotrees.
Keywords :
image coding; vector quantisation; wavelet transforms; embedded zerotrees; image data compression; oriented bands; separable dyadic transform; symmetry-adapted wavelet transform image compression; vector quantizer; wavelet decomposition; Computer science; Data compression; Filter bank; Fractals; Frequency; Image coding; Signal representations; Transform coding; Wavelet analysis; Wavelet transforms;
Conference_Titel :
Image Processing, 1999. ICIP 99. Proceedings. 1999 International Conference on
Conference_Location :
Kobe
Print_ISBN :
0-7803-5467-2
DOI :
10.1109/ICIP.1999.822896
