Images and image sequences analysis and coding based on rate distribution of embedded zerotree wavelet symbols

Abstract

Images and image sequences are the important contents of the multi-media information. As well for images as for image sequences, storage space or bandwidth must be large if there is no compression. After the discrete cosine transform (DCT) which is used in JPEG for still image compression, in MPEG for image sequences compression and in H263+ for low bit rate image sequences coding, wavelet transform (WT) is the new technology used now since it outperforms the DCT, in particular in still images compression. This is confirmed by JPEG2000 which is based on WT. The bloc effects which are the main drawback of DCT are overcome by WT technology. In this paper, we present the images and image sequences analysis based on the rate distribution of embedded zerotree wavelet (EZW) symbols in one hand and in the other hand its application to coding. The results are competitive with other codecs in the literature.