ONLINE HANDWRITING RECOGNITION SYSTEM FOR ENGLISH LANGUAGE USING SUPPORT VECTOR MACHINE (SVM) APPROACH

Abstract

The advances in the field of Artificial Intelligence open many gates in the direction of Human Computer Interaction (HCI). Many pattern recognition algorithms have been proposed for the applications where, interaction among machines tries to be simulated as Human-Human Interaction. One of these algorithms is handwriting recognition. The purpose is to convert the handwritten text to equivalent digital representation or simply typed form. Handwriting recognition applications falls in two categories: Online handwriting recognition and Offline handwriting recognition. In case of Offline part, the input has been provided as a scanned image containing handwritten text. But, in case of Online part, the input is provided in the form of digital ink collected with the help of digitizer and digital pen. In this paper, a system is proposed for Online handwriting recognition by considering English language and Roman script. Support Vector Machine (SVM) is used as a classifier. The complete word level classification has been performed and the recognition rate of 98.91% has been observed during experimentation. The technique can further be used for bilingual handwriting recognition systems.