Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/234505
Full metadata record
DC FieldValueLanguage
dc.coverage.spatial
dc.date.accessioned2019-03-26T08:53:18Z-
dc.date.available2019-03-26T08:53:18Z-
dc.identifier.urihttp://hdl.handle.net/10603/234505-
dc.description.abstractIn the current digital era, due to very heavy usage of computers,smartphones and computing devices in all aspect of human life, it is expected that information/document has to be in e-form. So, such devices should recognize native newlinelanguages to help common people to perform their daily tasks. Also, the paper is a newlinevery comfortable and feasible medium to store data, there is a great demand for the newlinesoftware techniques that can automatically extract, analyse and store information newlinefrom the physical handwritten documents for later retrieval. But communication newlinewith the computer using this natural means of paper is not possible due to high newlinevariation of handwriting and large character set for the Indian languages like Gujarati. Handwritten text recognition for the Indian scripts is still in the need of maturity for some of the languages like Gujarati. newline newlineRecognizing handwritten character of Indian scripts are more challenging due to its large character set and nature of character shapes. In the presented work we have formulated offline handwritten text recognition system capable of recognizing newlineisolated characters and text for Indian languages. The aim is to investigate the Handwritten text recognition for Gujarati script that can be generalized for Indian scripts. In this research, we have model the architectural solution which consisting of newlinepreprocessing, segmentation and recognition phases. Standard handwritten text newlinedocument dataset is not available for the experiment. We have built dataset consist newlineof 107 documents. newlinePreprocessing of image document is essential for reduction in data and for improving quality for better processing. Also, information in text document is best newlinerepresented as foreground and background. We have identified preprocessing phase newlinetask as converting image to grayscale, removing noise, binarization, cropping and newlineskew detection and correction. After the preprocessing text document image is newlinetransform into binary image which is then segmented in next phase. newline newline newline
dc.format.extentAll Pages
dc.language-1
dc.relation
dc.rightsuniversity
dc.titleDesign and Development of Text Line Segmentation and Recognition of Offline Handwritten Gujarati Text
dc.title.alternative
dc.creator.researcherNasriwala Jitendra Vinodchandra
dc.subject.keywordMachine Learning - Text recognization system
dc.description.noteComputer Science , Digital Image Processing
dc.contributor.guidePatel Bankim
dc.publisher.placeBarodli
dc.publisher.universityUka Tarsadia University
dc.publisher.institutionFaculty of Computer Science
dc.date.registered25/06/2012
dc.date.completed2017
dc.date.awarded11/12/2017
dc.format.dimensions
dc.format.accompanyingmaterialCD
dc.source.universityUniversity
dc.type.degreePh.D.
Appears in Departments:Faculty of Computer Science

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File40.58 kBAdobe PDFView/Open
02_certificate.pdf485.68 kBAdobe PDFView/Open
03_preliminary.pdf118.67 kBAdobe PDFView/Open
04_chapter 1.pdf256.02 kBAdobe PDFView/Open
05_chapter 2.pdf217.48 kBAdobe PDFView/Open
06_chapter 3.pdf284.66 kBAdobe PDFView/Open
07_chapter 4.pdf380.34 kBAdobe PDFView/Open
08_chapter 5.pdf520.43 kBAdobe PDFView/Open
09_chapter 6.pdf684.72 kBAdobe PDFView/Open
10_chapter 7.pdf366.29 kBAdobe PDFView/Open
11_chapter 8.pdf70.95 kBAdobe PDFView/Open
12_bibliography.pdf164.01 kBAdobe PDFView/Open
13_appendices.pdf292.81 kBAdobe PDFView/Open
14_publications.pdf66.83 kBAdobe PDFView/Open


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

Altmetric Badge: