ROI progressive image transmission based on wavelet transform and human visual specialties

Abstract

Based on the properties of integer wavelet transform and multiresolution analysis and combined with human visual specialties, an improved region of interest (ROI) progressive image transmission algorithm is proposed. With no impact on the subjective quality of ROI image, in the first instance, the coefficients in the ROI and the important wavelet coefficients for human vision are transmitted. Moreover, to indicate the relative importance of the BG with respect to the ROI, an expansion factor S for background (BG) image transmission and its formula is presented. Depending on the factor, the transmission of BG image can be controlled at different bit rates, so that the users can achieve the image with better subjective quality. Meanwhile, the algorithm could improve the visual quality of image and enhance the efficiency of the image transmission. The experimental results prove that the proposed algorithm is effective and the expansion factor could control the image transmission.