Content area

Abstract

The principal goal of this dissertation is to design image coders based on the two channel conjugate vector quantization. First the concept of two channel conjugate vector quantization is introduced, then its application to digital images is investigated. Image coders based on two channel conjugate vector quantization are developed in both spatial and transform domains. The emphasis here is on the usefulness of the two channel conjugate vector quantization when it is combined with conventional image coding techniques in both spatial and transform domains.

In spatial domain, the performances of the two channel conjugate vector quantizations are similar with those of ordinary VQs for noise free channel. However, when channel noise is injected, for the same bit rate, images reconstructed based on the two channel conjugate vector quantization are subjectively more pleasing (less visible distortion) compared to those based on ordinary VQ.

In transform domain, an adaptive image coding scheme, called classified discrete cosine transform vector quantization, is proposed to efficiently exploit correlation in large image blocks by taking advantage of transform coding (TC) and vector quantization (VQ), while overcoming the suboptimalities of TC and avoiding the complexity of VQ. Discrete cosine transform (DCT) is used here because of its superior energy compaction property with relatively moderate computational complexity. Also the two channel conjugate vector quantization is applied to the classified discrete cosine transform. For noise free channel, the performances of both techniques are similar. But, when channel error is injected, for the same bit rate, two channel conjugate classified discrete cosine transform vector quantization gives subjectively much more pleasing (less visible distortion) images than the classified transform vector quantization which is based on ordinary VQ. Computer simulations demonstrate the superiority of the two channel conjugate vector quantization technique for image data compression at various bit rates in both spatial and transform domains.

Details

1010268
Title
Image coding based on two-channel conjugate vector quantization
Number of pages
162
Degree date
1991
School code
2502
Source
DAI-B 52/10, Dissertation Abstracts International
ISBN
979-8-207-70175-2
University/institution
The University of Texas at Arlington
University location
United States -- Texas
Degree
Ph.D.
Source type
Dissertation or Thesis
Language
English
Document type
Dissertation/Thesis
Dissertation/thesis number
9208293
ProQuest document ID
303994884
Document URL
https://www.proquest.com/dissertations-theses/image-coding-based-on-two-channel-conjugate/docview/303994884/se-2?accountid=208611
Copyright
Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.
Database
ProQuest One Academic