In this paper, we propose a novel heuristic approach to segment recognizable symbols from online Kannada word data and perform recognition of entire word. Two different estimates of 1st derivative are extracted from the preprocessed stroke groups and used as features for classification. Estimate 2 proved better with 88% accuracy, which is 3% more than that achieved with estimate 1. Classification is performed by the Statistical DTW (SDTW) classifier which uses X, Y co-ordinates and their 1st derivatives as features. Classifier is trained with 40 writer data making it writer independent. 295 classes are handled covering Kannada aksharas, with Kannada numerals, Indo-Arabic numerals, punctuations and other special symbols like $, # etc. Akshara level classification accuracy is 88% and a lower accuracy of 80% at word level, which shows the scope for further improvement in segmentation algorithm.
Added on August 14, 2014
Product Type : Research Paper
License Type : Freeware
System Requirement :
Author : Rituraj Kunwar, Shashikiran K, A. G. Ramakrishnan