Please use this identifier to cite or link to this item:
Title: Design of vector quantization techniques for transform coding of images with orthogonal polynomials
Researcher: Kannan, N
Guide(s): Krishnamoorthi, R
Keywords: vector quantization techniques
orthogonal polynomials
Computer Science
Upload Date: 16-Oct-2012
University: Bharathidasan University
Completed Date: February, 2009
Abstract: The ever growing need for the transmission of digital images has urged the necessity for the design of efficient image compression techniques to reduce the data rate. The two main categories of image compression are lossless and lossy. Lossy image compression is based on compromising the accuracy of the recovered image in exchange for more compression. Generally lossy image compression involves a transformation to compact most of the energy of the input image into a few transform coefficients which are then quantized and entropy coded. The quantization process which represents a large set of values with a much smaller set plays a major role in lossy compression techniques. Quantization can be broadly classified into two types, namely Scalar Quantization (SQ) and Vector Quantization (VQ). An SQ involves mapping of each individual input to an output whereas VQ involves mapping a group of inputs into a set of well defined vectors using some distortion measure. The objective of this research is to design various vector quantization techniques with a combined approach utilizing both orthogonal polynomials based transform coding and vector quantization techniques that can give better compression ratio with good quality of reconstructed image. The proposed transformed vector quantization techniques are cored on a new orthogonal polynomials model. This model is based on a class of orthogonal polynomials which are used for obtaining point spread operator for different sizes of image regions. Any image region can be considered to be a linear combination of the responses of a complete set of these difference operators which are constructed from the point spread operator. After designing the orthogonal polynomials model, a new integer image coding technique with scalar quantization is devised with the proposed model. A scheme to find the orthogonal polynomials transform coefficients with high energy compaction is proposed for truncating the low energy orthogonal polynomials transform coefficients.
Pagination: 132p.
Appears in Departments:Department of Computer Science and Applications

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File69.32 kBAdobe PDFView/Open
02_certificate.pdf44.5 kBAdobe PDFView/Open
03_declaration.pdf55.4 kBAdobe PDFView/Open
04_publications.pdf49.26 kBAdobe PDFView/Open
05_acknowledgements.pdf7.33 kBAdobe PDFView/Open
06_abstract.pdf9.22 kBAdobe PDFView/Open
07_contents.pdf43.95 kBAdobe PDFView/Open
08_figures.pdf41.27 kBAdobe PDFView/Open
09_tables.pdf35.1 kBAdobe PDFView/Open
10_chapter 1.pdf96.27 kBAdobe PDFView/Open
11_chapter 2.pdf318.43 kBAdobe PDFView/Open
12_chapter 3.pdf295.53 kBAdobe PDFView/Open
13_chapter 4.pdf310.7 kBAdobe PDFView/Open
14_chapter 5.pdf255.51 kBAdobe PDFView/Open
15_chapter 6.pdf205.89 kBAdobe PDFView/Open
16_chapter 7.pdf25.44 kBAdobe PDFView/Open
17_references.pdf106.14 kBAdobe PDFView/Open

Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.