In this paper, we report the results of recognition of online handwritten Tamil characters. We experimented with two deferent approaches. One is subspace based method wherein the interactions between the features in the feature spate are assumed to be linear. In the second approach, we investigated an elastic matching technique using dynamic programming principles. We compare the methods to find their suitability for an on-line form-filling application in writer dependent, independent and adaptive scenarios. The comparison is in terms of average recognition accuracy and the number of training samples required obtaining an acceptable performance. While the first criterion evaluates elective recognition capability of a scheme, the second one is important for studying the electiveness of a scheme in real time applications. We also perform error analysis to determine the advisability of combining the classifiers.
Added on September 8, 2014
Product Type : Research Paper
License Type : Freeware
System Requirement :
Author : Niranjan Joshi , G. Sita, A.G. Ramakrishnan, Sriganesh Madhvanath