Based On Wavelet Image Compression Technology

The wavelet transformation can provide a good mechanism to explore human visual system, while it removes the data correlation in image. So the image compression coding algorithm based on wavelet transformation has became researching hotspot in recent years. In this dissertation, a image compression coding method based on wavelet transformation and vector quantization is discussed in detail.Firstly, the general conception of image compression is presented, the brief developing history of image compression technique is reviewed and several image compression international standards are introduced briefly. Then, the conception of wavelet transformation is introduced from the viewpoint of signal processing, and its application in image compression, including the selection of wavelet basis, the confirmation of decomposition stages and the distribution of wavelet coefficients, is researched by the experimental ways. In quantization stage, the lattice vector quantization is discussed firstly. Then a successive approximation wavelet vector quantization(SA-W-VQ) algorithm, which based on lattice vector quantization, is discussed in detail. Finally, according to the characteristic of wavelet coefficient distribution and human visual system, two improvements on this algorithm are presented . In entropy coding stage, the Huffman coding and the arithmetic coding is introduced, and the implementation of adaptive arithmetic coder is emphasized. The predication coding which used to code the lowest frequency wavelet subband losslessly is also discussed .In the last section of this dissertation, the simulation results for SA-W-VQ algorithm and the improved algorithm are presented. The simulation results show that compared to the original SA-W-VQ algorithm, the improved algorithm can obtain better results in both objective and subjective image quality, the PSNR increased up to 0.24 dB.