Feature Based Kannada Character Classification Method of Kannada Character Recognition

In Kannada language there are more than 510 character sets to be recognized including the vowels, consonants, consonant modified by vowel and consonant conjuncts the classification of characters becomes difficult. In this paper we present a new classification method of Kannada characters which can be used as a preliminary step for recognition. An analysis of Kannada characters was done and syntactic features were identified. Firstly the basic features present in the vowels are identified in the exact position of the character and recorded. Then by using a decision tree the characters are classified. The experimental results show the syntactic based method using basic features gives high contribution and reliability for Kannada character classification. Index words—vowels, consonants, vowel modifiers, consonant modifiers, consonant conjuncts, decision trees