Multiwavelet Preprocessing Methods And Their Application Of Vector Quantization

On the base of summarizing the multiwavelet theory the thesis studied several preflltering methods before decomposited the image by multiwavelet at large and analysised the relationship between multiwavelet and preflltering methods from several points of view.Simulation experiments proved that using preflltering method can greatly improve the image's energy centrality than not using prefiltering.And different methods brings out different effects to specified multiwavelet,which explained that preflltering is connected with specified multiwavelet.Next,the author brought forward a new method based on above study,with several experiments validating this conclusion by computer.As application of this new preflltering method,the thesis connected the multiwavelet decompositing image with vector quantization.It quantizated images that decomposited by GHM multiwavelet using three preflltering methods and testified that image with excellent energy centrality can get greater PSNR after quantization and restruction.