A dynamic codebook parallel vector quantizer for image compression and transmission

Gregory, M 1991, A dynamic codebook parallel vector quantizer for image compression and transmission, Masters by Research, Department of Communication and Electrical Engineering, RMIT University.


Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Gregory.pdf Thesis application/pdf 8.80MB
Title A dynamic codebook parallel vector quantizer for image compression and transmission
Author(s) Gregory, M
Year 1991
Abstract This document details research into the development of a Parallel Vector Quantiser using a Dynamic Codebook. To enable this objective to be reached the research has incorporated the development of a “real time” simulation evaluation system for such a quantiser. The complexities of modern parallel coding techniques make it such that direct comparison of new techniques with existing techniques is a time and facility consuming task that can often be inconclusive due to the use of different hardware and software. The simulation evaluation system has been developed to provide a platform for the efficient implementation of different parallel coding techniques and to permit conclusive comparison of the cost, complexity and fidelity retention of a parallel processing technique. The system is based upon a multiprocessor/shared memory architecture and the system‟s software kernel has been developed using a Graphical User Interface to permit the operator to simultaneously view the data stream “before” coding and “after” decoding in different display windows. The simulation evaluation system was used to develop and validate a Waveform Vector Quantiser using a Dynamic Codebook. Comparative results with other parallel coding techniques, based upon the use of a predetermined communication bandwidth, are presented and show that this technique produces higher fidelity with a slight computational and hardware complexity cost.
Degree Masters by Research
Institution RMIT University
School, Department or Centre Department of Communication and Electrical Engineering
Keyword(s) vector quantization
video coding
parrellel
Versions
Version Filter Type
Access Statistics: 481 Abstract Views, 112 File Downloads  -  Detailed Statistics
Created: Thu, 16 Jun 2011, 10:14:54 EST
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us