Design of Multimedia Application for Fast and Efficient Text Input from Touch Screen Input Devices using Character Recognition
K S Jagadeesh1, Chandramouli.H2, Naveen Ghorpade3
1K S Jagadeesh, Research Scholar, JJT University, Rajasthan.
2Chandramouli. Research Scholar, JJT University, Rajasthan.
3Naveen Ghorpade, Assistant Professor, Department of CSE, Sri Krishna Institute of Technology, Bengaluru.
Manuscript received on November 02, 2012. | Revised Manuscript received on November 26, 2012. | Manuscript published on December 30, 2012. | PP: 39-42 | Volume-2, Issue-2, December 2012. | Retrieval Number: B0827112212 /2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: We are wasting a lot of our time texting and typing messages through mobile’s and keyboards, so we have come up with software which can recognize the set of character u scribble on the screen and make it visible in the normal times new roman format. This would save lot of our time as we write or scribble faster than typing through other input devices and more efficient user interface is also achieved. Character recognition is a task of determining handwritten characters /digits. This is done by having some of the sample sets of characters written by numerous people. The task entails matching the handwritten characters with characters in the sample set and determining the character in the sample set which best matches the Test Character. The aim of the second step of the recognition structure is to extract discriminant information from an image of a character, as well as to reduce its dimensions of representation. This reduction is required in order to make easier the conception of the classification system, when discriminant feature extraction allows to present competently a character to the classifier. This paper envisages using a number of benchmark datasets to carry out the task. The first step is a feature extraction. Features such as shape, orientation, outline, character frontiers etc, have to be extracted from the character image. The features are then used for the pattern classification task. The output gives the class to which the character belongs. The results obtained using neural networks was compared with other methods of classification for character recognition and classification provides highest accuracy of 96%.
Keywords: Feature extraction, Transducer, Character Recognition, Pattern Recognition.