Applications of Stochastic Models for Image Data Compression.

Abstract : Intraframe image data compression systems are analyzed in this thesis using stochastic modeling concepts. The formulations of stochastic image models are obtained from different classes of partial differential equations. Their application to the coding problem shows the connection between predictive, hybrid and transform coding schemes. The resulting coding schemes are evaluated in terms of tangible system terms such as signal to noise ratio, mean square error and rate distortion curve. (Author)