Offline Tamil Handwritten Character Recognition using Zone based Hybrid Feature Extraction Technique
L. Anlo Safi, K. G. Srinivasagan · 2013
Character recognition is the most important research area in today’s world. Many researchers have focused on recognizing handwritten digits, numerals and characters in so many languages. To the best of our knowledge, little work has been done in the area of Tamil handwritten character recognition but they did not achieve better accuracy. Feature extraction is the important phase of character recognition. Feature extraction increases the recognition accuracy. This paper presents an overview of Feature Extraction techniques for offline recognition of Tamil characters. The proposed method is Zone based hybrid approach for feature extraction. The 55 features which are extracted from the character image are the number of horizontal, vertical, diagonal lines along with their total length for each zone. ANN classifier is used for classification and recognition purpose. We obtained a better accuracy when compared with the previous approaches for offline Tamil handwritten character recognition.