2002 · 21 citations · 5 references
ComputingLinguisticsNatural Language ProcessingPapermakingSpelling InstructionHidden Markov ModelsPattern RecognitionCharacter ClassificationHandwriting RecognitionHandwritten Character RecognitionStructural Properties Of LanguagesText Line Detection
This paper describes a method to recognize off-line handwritten Sinhala characters, the language used by the majority of Sri Lanka. The classification approach is based on discrete hidden Markov models. A subset of the Sinhala alphabet was chosen for the study. The unknown characters are first pre-classified into one of three character groups, based on the structural properties of the text line. This resulted in 99.9% accuracy. The HMM classifier is then used for the final recognition. The system was trained using 750 handwritten character images. A separate set of 500 character images was used to test the system. All the characters were written by 5 different writers on a preformatted paper. Results have shown 64.3% recognition rate for the first choice and 92.1% recognition rate up to the third choice.
5