Real Time Implementation of Telugu Character Recognition using Raspberry Pi

The market for small hand-held devices like mobiles devices and tablet computers is constantly growing. User interfaces designed in native languages of the users, can increase the reach and usability of these devices to wider population. Till now there are many tools available for English language but there are only few for Telugu language. This paper describes the implementation of Telugu character recognizer (CR) using Raspberry pi board. The hardware which is used here in order to develop online character recognition system is Raspberry Pi which has ARM11 processor. Users try to write the letters using mouse and it is displayed on the LCD screen. Two algorithms K-NN (nearest neighbor) and DTW (dynamic time wrapping) are used to recognize the letter. By using these algorithms and based on the hardware used, the text output and the voice output for the particular letters are generated. For Disabled people who can't speak and hear, the only way of communication for them is through Writing. Using these type of character recognizer tools helps them to communicate easily with the devices. This paper develops an user interface in native languages, like Telugu Language which helps elder people who are more familiar with their native languages to interact with devices in a more effective and easy manner and also kids can make use of this character recognizer tool to learn native languages easily.