Abstract

This thesis presents a new image coding using wavelet transform and adaptive block truncation coding. Images are first pre-processed by the wavelet transform and then coded by the adaptive block truncation coding. Algorithms for both monochrome and color images are proposed and experimentally studied. The adaptive block truncation coding is also modified to achieve better performance. For coding monochrome images at the bit-rate region between 0.8 to 1.2 bits/pixel, the performance of the new coding is comparable to the ones of subband codings and other image codings using the wavelet transform; however, the new coding offers less computational load. The new coding also gives a good reconstruction of a color image at the bit-rate of 1.0 bit/pixel. The comparison between the new coding and the original adaptive block truncation coding is also given. The discussion on effects of a filter and a number of decomposition levels used for an implementation of the wavelet transform is included in this thesis, as well.

Library of Congress Subject Headings

Image processing; Coding theory

Publication Date

6-1-1991

Document Type

Thesis

Department, Program, or Center

Electrical Engineering (KGCOE)

Advisor

Raghuveer, M.

Advisor/Committee Member

Salem, E.

Advisor/Committee Member

Mathew, A.

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works. Physical copy available through RIT's The Wallace Library at: TA1632 .C46 1991

Campus

RIT – Main Campus

Download

Share

COinS