Please use this identifier to cite or link to this item:
Title: Comprehensive Study of the Techniques Used For Online Handwriting Character Recognition
Authors: Kaur, Ramanjeet
Supervisor: Kumar, Rajiv
Keywords: Stroke, Pen Down, Pen Up, Primitives, Dominant Points
Issue Date: 25-Aug-2009
Abstract: Online automatic recognition of handwritten text has been an on going research problem for four decades. It has been gaining more interest lately due to the increasing popularity of hand held computers, digital note-books and advanced cellular phones, more accurate electronic tablets, more compact and powerful computers, and better recognition methods. In Online system input is taken as the (x,y) co-ordinates of the sampled points along with the time information. One can extract the order of the writing (where did the writer start, and where did he or she stop writing, what direction did the pen take) and also the speed of the writing. The first part of my work gives an introduction to handwriting recognition. The topics considered include; types of handwriting systems; difference between Online and Offline handwriting recognition; three different approaches to online handwriting recognition; online handwriting recognition system; and finally literature survey. Second part explains the online handwriting recognition method based on dominant points in strokes. This technique is based on sequential handwriting signals. In this approach, an online handwritten character is characterized by a sequence of dominant points in strokes and a sequence of writing directions between consecutive dominant points. The directional information is used for character pre-classification and positional information is used for fine classification. Both pre-classification and fine classifications are based on dynamic programming matching using the idea of band limited time warping. Third section discusses about the prototype (template) based online handwriting recognition method. An online handwriting system must be able to recognize a wide variety of handwriting styles, while attempting to obtain a high degree of accuracy when recognizing data from any one of those styles. As the number of writing styles increases, so does the variability of the data’s distribution. We then have an optimization problem: how to best model the data, while keeping the representation as simple as possible? If we can identify N different styles of writing individual characters (referred to as lexemes), these can then be modeled as N relatively simple independent distributions. This technique used a string matching distance measure for the recognition of online handwriting which takes the advantage of lexemes to reduce the number of prototypes that must be stored. A method of lexeme representation is shown. Fourth part presents the online handwriting recognition using structural based method, in which structure of the input is extracted and finally matched with the structure of models already stored in a model database to determine the class of input character by using elastic structural matching. In the last, I have discussed comparative study on the basis of advantages and disadvantages of all the techniques presented in this work.
Appears in Collections:Masters Theses@SOM

Files in This Item:
File Description SizeFormat 
904.pdf2.01 MBAdobe PDFThumbnail

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