%0 Journal Article
%T COMPARISON OF CLASSIFIERS FOR GUJARATI NUMERAL RECOGNITION
%A BAHETI M.J.
%A KALE K.V.
%A JADHAV M.E.
%J International Journal of Machine Intelligence
%D 2011
%I Bioinfo Publications
%X Handwritten script has been the most acknowledged method of collecting, storing and transmitting information allthe way through the centuries. In this paper an attempt is made to compare the offline handwritten character recognitionsystem for the isolated Gujarati numerals. For feature extraction affine invariant moments based model is used. We are usingKNN classifier and PCA (to reduce dimensions of feature space) and used Euclidean similarity measure to classify thenumerals. KNN classifier yielded 90 % as recognition rate whereas PCA scored recognition rate of 84%. The comparison ofKNN and PCA is made and it can be seen that KNN classifier has shown better results as compared to PCA classifier
%K Affine Moment Invariant
%K PCA
%K KNN
%U http://www.bioinfo.in/uploadfiles/13252461513_3_17_IJMI%20.pdf