Handwritten Gujarati Character Recognition using Machine Learning Approach (Record no. 115917)
[ view plain ]
| 000 -LEADER | |
|---|---|
| fixed length control field | ngm a22 7a 4500 |
| 008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION | |
| fixed length control field | 190222b ||||| |||| 00| 0 eng d |
| 082 ## - DEWEY DECIMAL CLASSIFICATION NUMBER | |
| Classification number | TT000066 |
| Item number | SHA |
| 100 ## - MAIN ENTRY--PERSONAL NAME | |
| Personal name | Sharma, Ankit |
| 245 ## - TITLE STATEMENT | |
| Title | Handwritten Gujarati Character Recognition using Machine Learning Approach |
| Statement of responsibility, etc | by Ankit Sharma |
| 260 ## - PUBLICATION, DISTRIBUTION, ETC. (IMPRINT) | |
| Place of publication, distribution, etc | Ahmedabad |
| Name of publisher, distributor, etc | Nirma Institute of Technology |
| Date of publication, distribution, etc | 2017 |
| 300 ## - PHYSICAL DESCRIPTION | |
| Extent | 106p Ph. D. Thesis with Synopsis and CD |
| 500 ## - GENERAL NOTE | |
| General note | Guided by: Dr. Dipak Adhyaru and Dr. Tanish Zaveri With Synopsis and CD 12EXTPHDE93<br/><br/>ABSTRACT:<br/>Handwritten character recognition is an active area of research. Over the past three<br/>decades, there has been increasing interest among researchers in problem related to<br/>the machine simulation of the human reading process. Optical Character Recognition<br/>(OCR) is the tool that is utilized to convert printed or handwritten scanned document<br/>into machine readable form/text. Handwritten character recognition is a challenging<br/>task and people are striving to convert handwritten literature to computer readable<br/>format. Recognising handwritten characters is dicult compared to printed charac-<br/>ters because handwritten characters may vary from person to person with respect to<br/>the individual writing style, size, curve, strokes and thickness of characters.<br/>Languages have played a major role in Indian history and they continue to<br/>in<br/>uence the lives of the Indians till date. Plentiful research on OCR techniques for<br/>Indian languages such as Hindi, Tamil, Bangla, Kannada, Gurumukhi and Malayalam<br/>has already been carried out. Development of OCR systems for Gujarati script is<br/>still in infancy and hence, there exists many unaddressed challenging problems for<br/>research community in this domain. This clearly necessitates the need to attend the<br/>task of handwritten Gujarati character recognition. This thesis addresses the issues<br/>of handwritten Gujarati character recognition.<br/>Gujarati is the mother tongue of people belong to Gujarat state in India. All over<br/>the world more than 65 million people use Gujarati language for their communication<br/>purpose. As Gujarat is one of the eminent state of India, Gujarati is a well-known<br/>and culturally rich language. Gujarati Character Recognition oers more diculties<br/>like the most other Indian languages relative to the western languages due to these<br/>reasons: (a) number of classes are higher, (b) structure of characters in Gujarati script<br/>contains curves, holes and strokes which result in signicant variations in writing style<br/>of different persons, (c) presence of similar looking characters (d) unavailability of standard dataset for experimentation and validation.<br/>One of the signicant contributions of proposed work is towards the development<br/>of large and representative datasets for the task of recognising handwritten Gujarati<br/>characters and numerals. Benchmark datasets having 88,000 handwritten Gujarati<br/>character images and 14,000 handwritten Gujarati numeral images are developed.<br/>Special forms are utilized for dataset collection and isolated characters are extracted<br/>from these forms. Preprocessing steps including noise removal, size normalization,<br/>binarization and thinning are applied on each segmented numeral/character image.<br/>Systematic and exhaustive experiments are carried out on these developed datasets<br/>using dierent kinds of features and their fusion. Zone based, projection proles based<br/>and chain code based features are employed as individual features. It is also proposed<br/>to use the fusion of these features. Few novel features are also proposed to represent<br/>handwritten Gujarati characters. These features include features extracted based on<br/>structural decomposition, zone pattern matching and normalized cross correlation.<br/>Methods based on articial neural network (ANN), support vector machine (SVM)<br/>and naive Bayes (NB) classier are used for handwritten Gujarati character and<br/>numeral recognition. In case of individual features, chain code based features provided<br/>higher recognition accuracy values compared to other features which were 99.25% and<br/>99.47% with polynomial SVM for numerals and characters datasets respectively. In<br/>case of fusion based features, fusion of chain code based and zoning based features<br/>provided best results compared to other fusion based features. Proposed structural<br/>decomposition based features provided highest accuracy of 99.48% with polynomial<br/>SVM for handwritten characters. Experimental results show signicant improvement<br/>over state-of-the-art and validate our proposals. |
| 856 ## - ELECTRONIC LOCATION AND ACCESS | |
| Uniform Resource Identifier | http://repository.nirmauni.ac.in/jspui/handle/123456789/8356 |
| Public note | Institute Repository (Campus Access) |
| 942 ## - ADDED ENTRY ELEMENTS (KOHA) | |
| Source of classification or shelving scheme | Dewey Decimal Classification |
| Koha item type | Thesis |
No items available.