Discrete wavelet transform based adaptive image coding schemes for continuous tone images

Abstract

This Thesis addresses the issue of selecting a wavelet transform newlinefor compressing different types of images Various adaptive methods are newlinepresented in this thesis which explore the local information for improved newlineperformance This Thesis aims to propose new algorithms for wavelet based newlineimage compression of continuous tone still images to achieve high PSNR newlineNew algorithms have been proposed by combining discrete wavelet newlinetransform wavelet packet transform adaptive lifting scheme and particle newlineswarm optimization technique The proposed algorithms offer improved newlinePSNR compression ratio and better visual quality of the reconstructed image newlineThe proposed methodutilize one of the following concepts one being the newlineproperty of wavelet transform in which most of the detail sub band newlinecoefficients remains zero next the use of wavelet packets which pin point newlinesignal components present locally in frequency domain another based on newlinelifting schemes which are highly flexible for building new wavelet newlinedecomposition from existing ones and Particle Swarm Optimization newlinetechnique which is a modern heuristic algorithm applied to image newlinecompression newline newline