Handwritten Online Character Recognition for Single Stroke Kannada Characters

The objective of the work is to develop a database for Online Handwritten Kannada Characters and recognize single stroke Online Handwritten Kannada Characters (OHKC). The iball 5540U Pen Tablet is used to collect the handwritten character samples and to build the database. These samples are collected from people who are native to Kannada language. These samples are pre-processed and features are extracted. The pre-processing techniques include smoothing, resampling and normalization. The features such as normalized coordinates, normalized trajectory, normalized deviation is extracted. The recognition is carried out by K Nearest Neighbor (KNN) classifier and obtained average recognition rate of 66.08 and 73.47% for K=1 and K=3 respectively. Online handwriting recognition has wide applications such as form filling, word processing and signature verification.