Date of Award

1999

Degree Type

Thesis

Degree Name

Bachelor of Engineering Honours

Faculty

Faculty of Communications, Health and Science

First Advisor

Dr Hon Nin Cheung

Second Advisor

Ken Ang Li-Minn

Abstract

Digital Imaging has had an enormous impact on industrial applications such as the Internet and video-phone systems. However, demand for industrial applications is growing enormously. In particular, internet application users are, growing at a near exponential rate. The sharp increase in applications using digital images has caused much emphasis on the fields of image coding, storage, processing and communications. New techniques are continuously developed with the main aim of increasing efficiency. Image coding is in particular a field of great commercial interest. A digital image requires a large amount of data to be created. This large amount of data causes many problems when storing, transmitting or processing the image. Reducing the amount of data that can be used to represent an image is the main objective of image coding. Since the main objective is to reduce the amount of data that represents an image, various techniques have been developed and are continuously developed to increase efficiency. The JPEG image coding standard has enjoyed widespread acceptance, and the industry continues to explore its various implementation issues. However, recent research indicates multiresolution based image coding is a far superior alternative.

A recent development in the field of image coding is the use of Embedded Zerotree Wavelet (EZW) as the technique to achieve image compression. One of The aims of this theses is to explain how this technique is superior to other current coding standards. It will be seen that an essential part orthis method of image coding is the use of multi resolution analysis, a subband system whereby the subbands arc logarithmically spaced in frequency and represent an octave band decomposition. The block structure that implements this function is termed the two dimensional Discrete Wavelet Transform (2D-DWT). The 20 DWT is achieved by several architectures and these are analysed in order to choose the best suitable architecture for the EZW coder. Finally, this architecture is implemented and verified using the Synopsys Behavioural Compiler and recommendations are made based on experimental findings.

Share

 
COinS