Volume : VI, Issue : IV, April - 2017

Handwritten Recognition of Tamil Scripts Using HMM With Fuzzy Logic

L. G. X. Agnel Livingston, L. M. Merlin Livingston

Abstract :

 Ancestors recorded and used the information in various ways of documents. The handwritten text decayed over a period of the time. It is very difficult to preserve them in the same form. It is a need to preserve these olden valuable data’s and they should be converted to computerized Unicode. In this paper the Hidden Markov Model is combined with the fuzzy approach in recognition of Handwritten Tamil characters. Identified unknown characters are classified as one among the prototype characters using features distance from the frame and a suitable membership function. The unknown and prototype characters are pre–processed and considered for recognition. The algorithm is tested for about 750 samples of Tamil characters and the success rate obtained varies from 89% to 93%.

 

Keywords :

Article: Download PDF   DOI : 10.36106/ijsr  

Cite This Article:

L.G.X.Agnel Livingston, L.M.Merlin Livingston, Handwritten Recognition of Tamil Scripts Using HMM With Fuzzy Logic, INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH : VOLUME-6 | ISSUE-4 | APRIL‾2017

Number of Downloads : 566

References :