%0 Journal Article
%T Dataset Generation for OCR
%A Aparna Vara Lakshmi Vemuri
%J International Journal of Computer Trends and Technology
%D 2011
%I Seventh Sense Research Group
%X In this paper we propose feature extraction for Telugu handwritten recognition based on the candidate search and elimination technique. The initial candidates for recognition are found by applying by zoning method on input glyphs. We propose cavities as a structural approach suited specifically for Telugu script, where cavity vectors are used to prune the candidates by zoning. It gives the 100% features and cavity features of the input dataset
%K Telugu is one of the prominent scripts in India and Asia
%K with more than 62 million speakers
%U http://www.ijcttjournal.org/volume-2/issue-1/IJCTT-V2I1P111.pdf