Research Of Region Of Interest Coding Based On DWT And HVS

With the rapid development of multimedia technology and Internet communicationtechnology, the image compression is becoming an essential step in the multimedia technicalinformation, and the international has formulated a series of image compression standard. Inorder to solve the conflict between image compression quality and ratio, t he JPEG2000providesa new technology: image with a region of interest (ROI) compression coding. At present, thistechnology has become the research focus of the image compression coding area.The paper firstly introduces the principles and main algorithms of image compressiontechnology based on wavelet transmission; then, the paper analyzes the characteristics of thehuman visual system (HVS) processes image information, establishes the HVS model for imagecompression encoding, and confirms it’s effectiveness through experiments; finally, the paperstudies the ROI image compression technology based on the above. In this paper, the research ofROI as follows:1. Based on the research of the methods of JPEG2000’s ROI image coding and thoseimprovements, the paper studies the factors in ROI image coding. Through several experiment,we confirm that the wavelet coefficients are the importance of the whole image. The experimentupgrades the wavelet decomposition low-frequency and partial high-frequency coefficients bythe general scaling method of JPEG2000, in order to improve the quality of interested regionsand the whole images, and then code combining with SPIHT. Through three sets of contrastexperiments, we can reach this conclusion: reconstructed image quality has been increased aftercompletely improving low-frequency and intermediate frequency coefficients.2. For the texture-rich of ROI image, the paper presents an effective compression scheme,namely a region of interest based on human visual system image compression. Firstly, weweights coefficients of the image’s wavelet domain with different frequency and differentdirection. Secondly, we use the maxshift method to upgrade the ROI coefficients in the waveletdomain. Finally, coding is implemented with SPIHT algorithm. After coding for large amount ofimage data, the results show that this scheme has good invisibility to the ROI image with the richtexture.Finally, we give a summary and outlook.