Devanagari Letters Segmentation and Recognition System: A Brief Review

Character Recognition has been a significant area of study in Artificial Intelligence. The concept of Character Recognition became famous in recent times due to its many applications such as in printed postal addressing, filling of a variety of forms, multiple choice questions in certain examination and so on. OCR suggests to the technique of photo scanning of the text character-by-character, analysis of the scanned picture and then changing of the character image into information that a machine can easily read. e.g. organizations and libraries taking physical replicas of books, magazines, or other old printed material and utilizing OCR to place them into computers. Segmentation is the crucial and most complex part of OCR process, and it gets more difficult with handwritten text because of different writing styles and fonts used. Devanagari character recognition is more complex as it is having various loops, conjuncts, upper and lower modifiers and the number of disconnected and multistroke characters. This paper presents a comparative analysis of various character recognition techniques in terms of their accuracies and also their features, classifier and findings as well.