Wavelet zero-path mapping algorithm for multimedia images coding

In the age of information and communication technology (ICT), the requirements of multimedia image transmission are essential. Image compression (source coding) has provided significant improvements for communication bandwidth, and transmission cost and data rate. The objective of the paper is to develop new techniques for strengthening the compression of still images. The features of the proposed algorithm include: (i) a wavelet transform engine, which provides a better image compression rate and visual quality as compared to DCT-based systems (JPEG); (ii) zero-path which exploits the cross-scale self-similarity of wavelet coefficients and thus provides zero-path compression; (iii) the absence of side information (significant map) for coding and decoding of zero-tree structure hence reduces the data bits for significant map symbolic coding; (iv) adaptive arithmetic entropy coding provides further compression on generating variable-length codewords. Experimental results show that the proposed algorithm outperforms both JPEG and the EZW (embedded zerotree wavelet) algorithms in terms of peak signal-to-noise ratio (PSNR), data compression bit-rate (bpp), and also human visual perception.