Software for Tamil Handwritten Character Recognition (TCR)

Tamil is an Ancient and beautiful language and holds the prestigious stand of classical language declared by UNESCO. Integrating it with the technology is a try to enhance its beauty. Language integrating with technology has always proved to be a boon to the learners. There has been number of research done in integrating English language with technology and this has proved to be wider scope of research in the field of English language teaching itself. Tamil being the fifth most spoken language in India, holding the status of official language in Singapore and Sri Lanka apart from the recognition as minority language in South Africa Malaysia and Mauritius, deserves to be explored, enhanced and simplified for its learners at the very best. This Catered researchers interest towards integrating technology and Tamil language. Current scenario states the research study, of handwritten recognition of Tamil script is available only online. The purpose of this project is expanding its scope to make it available offline as well. Neural network algorithm which can be used to classify and recognize the pattern of language stands as a base to run this project. A Set of Sample Handwritten Tamil Characters are taken as input in the image format, to process the character, and train the Neural Network Algorithm, accordingly to recognize the pattern and convert recognized characters to a Printed document. The conversion of handwritten script to computerized text format is done. Neural Networks are particularly useful for solving problems that cannot be expressed as a series of steps, such as Recognizing patterns, Classifying them into groups, Series prediction and Data mining. The Segmentation involves spacing method and Feature Extraction is done using zoning method. The Neural Network then attempts to determine if the input data matches a pattern that the Neural Network has memorized. A Neural Network is designed to take input samples and classify them into groups as it is trained for classification. This project concerns with improving the accuracy for composite letters. The main aspect of the project is that researchers have worked on the converting the handwritten text to a computerized text format. This type is first of its attempt in Tamil language technology integration.