An Efficient Writing-Zone Identification Technique for Online Handwritten Gurmukhi Character Recognition

Abstract

Handwriting recognition is a technique that converts handwritten characters into machine processable format. Handwritten characters can either be presented to machine online or offline. Research in the area of online recognition of Indian languages has attracted many researchers in the past, whereas, a good amount of research has also been carried out for English, Chinese, Japanese and Korean languages. Headline and baseline are common features in most Indic languages which divide a character into three zones, namely, upper, middle and lower zones. Identification of headline and baseline is a major task for identification of strokes located in these three zones. A writing zone identification algorithm is proposed and tested in this text for online handwritten Gurmukhi character recognition. The problem of formation of different characters based on writing similar strokes in different zones, is solved by identification of the writing zone. A rule based approach has also been applied and tested for generation of characters from the set of recognized strokes. The stoke classifier has been trained using Hidden Markov Model for 74 stroke classes yielding a 5-fold cross validation accuracy of 95.3% . In this work, an accuracy of 97.7% has been achieved for writing zone identification and an accuracy of 88.4% has been achieved for character identification for Gurmukhi script. This accuracy has been achieved when the recognition engine was tested on the dataset of 4280 characters written by 10 users.