No abstract

In this paper, we describe a wr method for recognizing offline unconstrained h words/strings. Touching and overlapping are m handwriting recognition and to overcome thi holistic approach here. To handle various different individuals, at first, some pieces of pr are done on an input word. Next, Gabor ba computed on the processed word. These Gabo with other geometric features of the word imag an SVM classifier for recognition. In our experi have used 4270 samples (collected from the cla names) written in Tamil and obtained 86.36% re