Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/13664
Title: Tamil text image processing restoration and character recognition
Researcher: Selvakumar Raja S
Guide(s): Mala John
Keywords: Tamil, Image processing, restoration, character recognition, Expectation Maximization (EM), Discrete Wavelet Transform, Back Propagation Neural Network, Support Vector Machine
Upload Date: 5-Dec-2013
University: Anna University
Completed Date: 
Abstract: Tamil, one of the oldest languages in the world, is a South Indian language spoken widely in Tamil Nadu, one of the states of India. The significance of the thesis stems from the fact that the Tamil language is ancient and widespread, owning rich literature content. Realizing the research potential in Tamil Text image problem, this thesis is focused towards Tamil Text Image Restoration and Tamil Character Recognition. Processing of the Tamil Text image, like any other text image primarily involves two components: Restoration and Character recognition. This thesis addresses the problem of Tamil text image restoration and Tamil character recognition, which form the integral parts of Tamil text image processing. The issues related to Tamil character recognition could be either related to feature extraction or to classification. As both feature extraction and classification have equally important roles to play, this thesis addresses both of these problems. The initial part of the thesis is focused towards restoration of noisy Tamil Text Images. A spatially adaptive restoration algorithm based on Expectation Maximization (EM) operating in the wavelet transform domain is proposed. The second part is dedicated to Tamil character recognition in terms of feature extraction and classification. Two novel feature extraction methods viz. Slope method and Discrete Wavelet Transform (DWT) domain based method are proposed. The slope method makes use of identifying unique curves in the Tamil character set which can be effectively used for classification purpose. The feature vectors extracted from the proposed methods are fed as inputs to different classifiers viz. Back Propagation Neural Network (BPNN), Support Vector Machine (SVM), Decision Tree classifiers and their performance is experimentally validated. The feature extraction algorithms and the classifier algorithm presented herein prove to be promising candidates for Tamil character recognition. newline newline newline
Pagination: xix, 110
URI: http://hdl.handle.net/10603/13664
Appears in Departments:Faculty of Information and Communication Engineering

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File25.34 kBAdobe PDFView/Open
02_certificates.pdf128.51 kBAdobe PDFView/Open
03_abstract.pdf13.64 kBAdobe PDFView/Open
04_acknowledgement.pdf15.93 kBAdobe PDFView/Open
05_contents.pdf122.39 kBAdobe PDFView/Open
06_chapter 1.pdf146.63 kBAdobe PDFView/Open
07_chapter 2.pdf1.62 MBAdobe PDFView/Open
08_chapter 3.pdf110.3 kBAdobe PDFView/Open
09_chapter 4.pdf991.85 kBAdobe PDFView/Open
10_chapter 5.pdf107 kBAdobe PDFView/Open
11_references.pdf393.6 kBAdobe PDFView/Open
12_publications.pdf37.78 kBAdobe PDFView/Open
13_vitae.pdf13.13 kBAdobe PDFView/Open


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